Skip to main content

Full text of "The system of the stars"

See other formats

This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project 
to make the world's books discoverable online. 

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject 
to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books 
are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover. 

Marks, notations and other marginalia present in the original volume will appear in this file - a reminder of this book's long journey from the 
publisher to a library and finally to you. 

Usage guidelines 

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the 
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing this resource, we have taken steps to 
prevent abuse by commercial parties, including placing technical restrictions on automated querying. 

We also ask that you: 

+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for 
personal, non-commercial purposes. 

+ Refrain from automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine 
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the 
use of public domain materials for these purposes and may be able to help. 

+ Maintain attribution The Google "watermark" you see on each file is essential for informing people about this project and helping them find 
additional materials through Google Book Search. Please do not remove it. 

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just 
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other 
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of 
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner 
anywhere in the world. Copyright infringement liability can be quite severe. 

About Google Book Search 

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers 
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web 

at |http : //books . google . com/ 







rnixTKn by 




I , 







*Io vidi delle cose hdle 
Che porta *l ciel ' 



All riffhtf rfterttd 

Attron. Dept. 




FEBRUARY 84, 1890 


Sidereal science has a great future before it. The prospects 
of its advance are incalculable'; the possibilities of its develop- 
ment virtually infinite. No other branch of knowledge at- 
tracts efforts for its promotion, at once so wide-spread, so 
varied, and so enthusiastic ; and in no other is anticipation 
so continually outrun by the brilliant significance of the 
results achieved. 

For the due appreciation, however, of these results, some 
preliminary knowledge is required, and is possessed by few. 
To bring it within the reach of many is the object aimed at 
in the publication of the present volume. Astronomy is 
essentially a popular science. The general pubHc has an 
indefeasible right of access to its lofty halls, which it is 
the more important to keep cleared of unnecessary technical 
impediments, since the natural tendency of all sciences is to 
become specialised as they advance. But literary treatment 
is the foe of specialisation, and helps to secure, accordingly, 
the topics it is applied to, against being secluded front the 
interest and understanding of ordinarily educated men and 
women. Now, in the whole astonishing history of the human 
intellect, there is no more astonishing chapter than that con- 


cerned with the sidereal researches of the last quarter of a 
century. Nor can the resources of thought be more effectually 
widened, or its principles be more surely ennobled through the 
vision of a Higher Wisdom, than by rendering it, so far as 
possible, intelligible to all. 

The following pages then embody an attempt to combine, 
in a general survey, some definite particulars of knowledge 
regarding our sidereal surroundings. The plan pursued has 
been to instruct by illustrative examples, to select typical 
instances from each class of phenomena, dwelling upon them 
with sufScient detail to awaken interest and assist realisation, 
while avoiding the tediousness inseparable from exhaustive 
treatment. The statement of facts has been kept primarily 
in view ; but the more important efforts to interpret them 
have been noticed, and the difiBculties attending rival theories 
impartially pointed out. In developing the subject it seemed 
best to proceed from the particular to the general ; to start 
with describing the physical constitution of individual bodies, 
and, ascending by degrees through continually added com- 
plexities of mutual relationships, reach at last the crowning 
problem of the Construction of the Heavens. 

The writer gratefully acknowledges the assistance derived 
in the preparation of the present wgrk, from the kindness of 
Dr. Gill, H.M. Astronomer at the Cape, first and chiefly, in 
affording her an opportunity of observing the southern skies ; 
secondly, in reading over several of its chapters in manuscript. 
Her thanks are also due to Professor E. S. Holden, Director 
of the Lick Observatory, to Messrs. Burnham, Keeler, and 
Barnard of the same establishment, to Professor E. C. Picker- 
ing, Director of Harvard College observatory, to Dr. and Mrs. 
Huggins, Professors Lockyer, Vogel, Schonfeld, and others, 


for communications of great interest and value. To the 
generous assistance of Mr. Isaac Eoberts, of Crowborough, the 
author is indebted for the photographs of nebulae and clusters 
reproduced in the Plates ; others have been kindly furnished 
by Mr. Barnard; while permission to use photographs and 
drawings already published has, with the utmost courtesy, 
been accorded to her by Admiral Mouchez and the MM. 
Henry, by Professors Holden and Pickering, Dr. and Mrs. 
Huggins, Professor Vogel and Dr. Scheiner of Potsdam, and 
by Dr. Duner, Director of the Observatory of Upsala. In 
designing and selecting the illustrations, elucidation, rather 
than ornament, was aimed at; how far the end has been 
attained, must be left to the public to decide. 

LiOMDON : October^ 1890. 





Nmnber of Visible Stars — Designations — Magnitudes — Scintillation of 
Stars — Multitude of Telescopic Stars — Sidereal Natural History— Dis- 
coveries of Nebulffi — Their continuous observation desirable 



Star Catalogues — Location of Stars on the Sphere— Personal Equation — 
The Equatoreal Telescope and Micrometer — The Ileliometer— Stellar 
Light Analysis— Spectroscopic Determination of Movement— Stellar 
Photometry — Celestial Photography — Star Charting — Photographic 
Magnitudes — Nebular Photometry — Photographic Parallax — Refractors 
and Reflectors — Advantages of the Photographic Method 12 



The Stars as Suns — Spectra of White Stars- Ultra- Violet Hydrogen Lines • 
—Gradual Transition from Sirian to Solar Stars — Chemistry of Alde- 
baran— Spectrum of Arcturus— Temperatures of the Stars— Physical 
Distinctions . 35 



Significance of the Appearance of Flutinf^s in Stellar Spectra- Of two 
Distinct Kinds— Stars of the Third Type - Chemical Constitution of 



Betelgeux — Lockyer's Hj-pothesis— Spectrum of Mira— Physical nature 
of Third Type Stars— Stars of Fourth Spectral Type— Scarcity- 
Redness— Distribution 52 



Fifth Spectral Class— Variable Bright Lines in Stellar Spectra— Constitu- 
tion of Gaseous Stars - Wolf-Rayet Variety — Transition to Nebulic — 
Spectrum of Gaseous Nebulte— Of the Orion Nebula— Photographic 
Investigations— Spectrum of the Andromeda Nebula . . . . 6& 



Life-History of Stars — Advance of Condensation— Relationships of Side- 
real Classes — Meteoric Hypothesis— Status of Red Stars — Of the Sun 
— Stars past their prime — Indicated Line of Evolution — Pre-Nebular 
Theories 82 



Degrees of Stellar Variability— Classification of Variables— Recorded New 
Stars — Tycho*B Star — Nov® of Kepler and Anthebn — Spectrum of 
T Corona?— Nova Cygni— Stellar outbursts in Nebulae — Nova Andro- 
medes— Distribution and Origin of Temporary Stars . • . .95 



istribution of Periods — Mira Ceti — Range of Variability — Spectrum — 
U Orionis- Chi Cygni— U Geminorum— Eta ArgAs — Irregular Variables 
— Theories of Light-Change — Collision Hypothesis — Analogy of 
Sunspot Cycle— Periods and Colours of Variables .... 108 



Marked Distinction of short from long period Variables— Beta Lyrie and 
B Sagittae — Delta Cephei— Eta Aquilrc— Zeta Geminorum— Two 



rapidly changing Stars— Explanatory Hypotheses— Algol-type of Stars 
—System of Algol— Conditions of Eclipse in other Stars— Significant 
Pecoliarities— Distribution of Variable Stars 129 



Ptolemy's Red Stars— The colour of Sirius— Examples of Change — Diffi- 
culty of Colour-Determinations — Catalogues of Red Stars — Colour 
variables — Distribution of Red Stars— Colours of Double Stars — Con- 
trasted Tints — Permanence or Instability — Supposed Relation to 
Orbital Movement— Correspondence of Spectral Absorption . /l46 



Early Discoveries— Numbers— Apparent Proximity — Binary Systems— 
Castor — a Centauri— 61 Cygni — 70 Ophiuchi and i? CassiopeifiB — Obscure 
Satellites — System of Sirius — Orbit of Procyon — Criteria of Physical 
Connection — Common Proper Motion — Colour Contrasts — Relative 
Variability— Future Investigations 163 



Nature of Light-Changes in Compound Stars — Relative Variability — 
Satellite of 8 Cygni — Variable Double Stars with Sirian Spectra — 
Sympathy in Light-Change a test of Physical Connection — Importance 
of the Study of Variability in Stellar Systems 179 



Universal Prevalence of Law of Gravitation— Measurements of Double 
Stars— Apparent Orbits — Substitution of the Camera for the Micro- 
meter— Computation of Real Orbits — Masses of Binary Stars— Periods 
—Relative Movements— Spectroscopic Method — Equation of Light in 
Stellar Orbits— Mass-Brightness— Elliptic Paths 187 





Further Division of Doable Stars— Ternary Systenu— 7 AndromedA— 
Bigei — 40 Eridani — Castor — System of C Cancri — Obscare Component 
— System of C Urs®— Spectrographic Discoveries of Stellar Satellites 
—Quaternary Stars — t LyrsB — v Scorpii — Multiple Groups — Nebular 
Relationships — Orion Trapezium— Transition to Clusters . . . 205 



gcndary Importance of the Group— The Lost Pleiad— Real Populousness 
of the Community — Drift across the Sphere — Measurements of in- 
cluded Stars — Estimated Distance from the Eartli — Absolute Brilliancy 
— Mechanism of the Cluster — Photographed Spectra — Nebulffi ,in_ the 
Pleiades— A Miniature Sidereal System . . . T . . . 220 



Globular and Irregular Clusters— Reticulated Structure — Signs of Subdi- 
vision --Dominant Red and Double Stars — Cluster about «c Crucis — 
Pra>sepe— Dynamical Equilibrium of Globular Clusters— Radiated Ap- 
pearance— « Centauri and 47 Toucani— Dark Lanes— The Great Cluster - 
in Hercules— Nebular Affinities of Clusters 234 



Fantastic Variety — Nebular Classification— Nebulous Stars— Planetary 
Nebula)— The Owl Nebula — Helical Forms — Annular Nebulte— The 
Ring-Nebula in Lyra— Cometary and Spiral Nebul© — Double Nebulae — 
The Dumb-bell Nebula 261 



First Observations of the Andromeda Nebula— Bond's Canals — Roberts's 
Photographs — Constitution and Real Shape of the Nebula — Analogy 



with a Star-Group — Other Elliptical Nebulro— Irregular Nebul© — Great 
Looped Nebula— Orion Nebula — Structure Visible and Photographic - 
Vast Extent — Analogy with Pleiades — Nebula in Argo — Omega and 
Trifid Nebula»— Diffused Nebulosities 268 



NebulflB and Comets — Structural Besemblanoes — Type-Curve of Helical 
Nebulffi — Indications of Repulsive Action — Immobility of NebulsB — 
Satellites of Planetaries — Variability in Light— Disappearances and 
Fluctuations — Local Changes in Irregular Nebulffl — Electrical Illu- 
mination 285 



The Problem of S teHar Parallax— Perspective Effects of Earth's Annual 
Revolution — Futile Efforts at Detection — Parallaxes of Binary Stars — 
Differential Method— First Genuine ■ Results— Criteria of Vicinity — 
Probable Errors — Exigencies of the Work — Measurements at the Cape 
— Mean Parallaxes — Distance-Ratio — Measurements at Yale College — 
Photographic Determinations— General Conclusions .... 297 



Movement of the Sun among the Stars— Perspective Effects— Apex of the 
Sun's Way — Method of Least Squares — Argelander's Treatment of the 
Problem— Modification by Airy— L. Struve's Discussion— Rate of Solar 
Motion— Spectroscopic Determination— Orbit of the Sun— Secular 
Aberration of Light — Obliquity of the Sun*s Course — Residual Stellar 
Movements 319 



Difficulty of Ascertainment — Minuteness of Quantities concerned — Move- 
ment a Test of Vicinity— Motions and Magnitudes — Anomalous Results 
—Swift Stars — Radial Movements — Photographic Determinations- 
Runaway Stars — Star Drift — Partial Systems — Possible Associates of 
the Sun— Relation to Milky Way vt Stellar Revolutions . . .334 





Visual Aspect— Great Rift— Interruptions— Black Opemngs - Charts - 
Structural Complexity— Disc-Thfeory— Effects of Clustering Power- 
Stratum Theory— Annular or Spiral Formation— Difficulties— Remote- 
ness of Aggregated Stars— Superposed Galaxies 3.>2 



Nebulffi not external Galaxies— Proofs of the Fact— Magellanic Clouds- 
Peculiarities of Nebular Distribution— Avoidance of the Milky Way- 
Frequented by Clusters and Gaseous Nebulae- Inferred Differences of 
Constitution — Indirect Evidence of Motion in Nebulft— Unity of Stellar 
and Nebular Systems 368 



A Dynamical Problem— Finite Character of the Sidereal System— Photo- 
graphic Star Gauging— The Solar Cluster— Difficulties connected with 
its Organisation — Elements of Sidereal System computed by Maxwell 
Hall— Law of Particular Aggregation— Star Streams and Rings — Shape 
of the Visible Universe— Progress and Change— Conclusion . . 379 

APPENDIX.— Tables of Stellab Data 399 

INDEX 411 



r. Chart of the Pleiades^ . .- ^ . . ^ - Frotitispiece 
II. Olobuule Cluster in Hercules .... To face page 246 

ni. Great Nebula in Andromeda «> » 270 

rV. Elliptical Nebula in Ursa Major „ „ 274 

T. Photograph of the Great Nebula in Orion. Exposure 

81 minutes .►......„„ 279 

YI. The Saaie. Exposure 3^ hours. , ^ • ^ . „ „ 281 


figurb taoe 

1. Spectrum of Hydrogen in White Stars compared with the 

Spectrum of the Orion Nebula (Huggins) . . . .38 

2. Photographed Stellar Spectra (Htiggins) 41 

3. Photographed Stellar Spectra (Scheiner) 43 

4. Stellar Spectra of the Third Class: 1, a Orionis ; 2, a Herculis 

(Dun^) 54 

5. Spectrum of Betelgbux (a Orionis) Photographed at Harvard 

College 57 

6. Spectrum of Miba showing bright lines of Hydrogen and dare 

lines of Calcium (Pickering) 59 

7. Stellar Spectra of the Fourth Class: 1, 152 Schjellerup; 2, 19 

PisciUM (Dun&r) 62 

8. Spectra of Gaseous Stars compared with the Fourth Type 

Spectrum of 162 Schjellerup (Vogel) 73 

9. Photographic Spectrum of the Great Nebula in Orion (Huggins) 79 
10. Changes in the Spectrum of Nova Cyont (Vogel) .... 101 




11. D18TBIBUTIOX OF 171 PiBioDB OF Vabublk Stabs pbox Oobb*8 Retised 

Cataixkjuk (1888) 109 

12. Maxima of Miba tx Febbuabt 1885, and Jancabt 1886 . .111 

13. Two Types of MAZixrM of U Oeionobcii (KnoU) . . 115 

14. LiohtGurfe of if Abo^b, ldl0>1890 118 

15. Thb Coixibion-Theobt of Vabiable Stabb 123 

16. GuBVE OF Sun-spot Freqcenct, 1867-77 (Ellis) . . . .125 

17. Dutbibdtion of all thb Pbbioi>8 of Vabiablb Stabs unobb twenty 

DATS 130 

18. Distribution of periods undbb twenty days, bzcludino those of 

Alool Variables 130 

19. LioHT-cTRTE OF fi L^nx (ArgeUmder) 131 

20. LiOHT-cuBVE of B Sagittjb (1884) 132 

21. LioHT-cuRVB OF > Cephei 132 

22. LiaHT-cusFE of ii Aquils 132 

23. Lioht-curfb of ( Obminobum 133 

34. Lioht-cubves of U Ophicchi (No. 1), and R Musca (No. 2), the 

two tariables of shobtest known periods 184 

25. Algol during an Ecupse 138 

26. Minimum of S Cancbi 140 

27. Inverted Lioht-curve of 10 Saoittjb [Oare 1885) . . . . 143 

28. Spectra of the Component Sta&s of fi Cyoni {Huggins) . . . 161 

29. Four Double Stabs photooraphed at Paris (from Mouchei'^s ' Pho- 

tographic Astronomiqtte^} 190 

30. Orbits of the Components of y Virginis 197 

31. Orbits of the Components of a Centauri 198 

32. Orbits of Snuus and its Companion 198 

33. Stabs of the Trapezium .... 218 

34. Stab Cluster in Sobieski's Shield (M 11) photographed by Mb. 

£. £. Babna&d, at the Lick Obserfatoby 237 

35. Plan- Sketch of a Nebula [Vogel) 256 

36. Planetaby Nebula SEaoDTED into Helical Form {Holden) . . 258 

37. Annular Nebula in Aquabius {Holden) 260 

38. Spiral Nebula in Canes Venatici. From a Photograph taken al 

Heriny 263 

39. Spiral Nebula in Virgo. From a Photograph taken at Heriny . . 264 

40. Diagram of the Great Nebula in Andbomeda. From Bond^s Dratving 

in 1847 270 


riouBR pAaa 


42. Photoobaph of the Mhjct Wat n Saoxttabiub .273 

43. Indbz-Duobax to Stbuctcbbb photoobaphed ni the Oriov Nebula. 279 

44. Gboup of Synclinal Stbuctubes fbox a Photoobaph of the Cobona 

OF 1871. Dravm by Mr, W. H, Wesley 280 

45. Nebula in Leo (M 66) photoobaphed bt M. ton Oothabd . 286 

46. Nebula in Geuini (M 65) photoobaphed bt M. von Gothabd . . 286 

47. Type -CURVE of Heucal Nbbulb viewed in various positionb 

(Holden) 288 

48. The Eabth^s Motion in Space (Young) 331 

49. Displacement of Hydbooen-linb in the Spectbuu of Bioel . . 342 

50. Distbibution of 934 btabs within 1° of the Pole, aHOWiNo the 

ratio of numbebs to space fob each half biagnitude . . . 363 


Supply at p. 8, noU, line 2 from bottom : 

Spica - a Virginia. 

Add to list of Algol-variables at p. 136 : 

Name Dlacorerer Period Amount aiid ilunttion of cftange 

12 AntU» Paul, 1888 7h. 47m. 67 to 73 mag. in 4h. 60 m. (?) 

NoU to page 284. 

The spectrum of the trifid nebula, as obsenred by Mr. Keeler at Lick in 
1890, is continuoMS in the green and blue, with a brightening near the middle. 

NoU to page 290. 

Mr. Keeler has lately succeeded in measuring, with the great Lick refractor, 
the movements in line of sight of several planetary nebulae. See Ptiblicatunu 
Astr. Pacific Society, No. 11, 1890. 

Note to page 342. 

The velocities of recession ascribed to fi and t Orionis in 1888 can only be 
regarded as provisionally determined. With improved apparatus, evidence has 
been obtained at Potsdam of variability in the movement of Rigel, due doubtless 
to orbital revolution. 




When all the stars blaze out on a clear, moonless night, it 
seems as if it would be impossible to count them ; and yet it 
is seldom that more than 2,000 are visible together to the 
unaided eye. The number, however, depends very much 
upon climate and sharpness of sight. Argelander enumerated 
at Bonn, where rather more than eight-tenths of the sphere 
come successively into view, 8,237 stars.^ But of these no 
more than 2,000 could be, at any one time, above the horizon, 
and so many would not be visibly above it, owing to the 
quenching power of the air in its neighbourhood. Heis, at 
Miinster, saw 1,445 stars more than Argelander at Bonn;* 
Houzeau recorded 5,719 at Jamaica ; ^ Gould 10,649 at 
Cordoba in South America.* The discrepancies of these figures, 
setting aside the comparatively slight effect of the increased 
area of the heavens displayed in low latitudes, are due to the 
multitude of small stars always, it. might be said, hovering 
on the verge of visibility. If, indeed, the atmosphere could 
be wholly withdrawn, fully 25,000 stars would, according to 
a trustworthy estimate, become apparent to moderately good 

Our system of designating the stars has come down to us 

' Uranometria Nova, 1843. 

' Heis, De Mckgnitudine Numeroque Stellarum, p. 16, 1852. 

* UranomStrie Q&n^aU, Annales de rObaervatoire de Bruxelles, t. i. 1878. 

* Uranometria ArgenHna, 1879. 

* Backhouse, Journal Liverpool Aelr, Society, vol. yii. p. 226. 



from a hoar antiqaity. It is a highly incommodious one. ' The 
constellations/ Sir John Herschel remarks/ 'seem to have 
been almost purposely named and delineated to cause as 
much confusion and inconvenience as possible. Innumerable 
snakes twine through long and contorted areas of the heavens 
where no memory can follow them ; bears, lions, and fishes, 
large and small, northern and southern, confuse all nomen- 
clature.' And yet we could ill afford to dispense with the 
picturesque associations of a menagerie largely stocked from 
the banks of the Euphrates. The signs of the Zodiac, which 
are undoubtedly of Chaldean origin, embody legendary cycles 
of thought already, some four thousand years ago, the worse 
for wear and dilapidated by time. Homer and Hesiod were 
familiar with the Bear, Arcturus, and the Dog-star, with 
' the Hyades, and the Pleiades, and the strength of Orion.' 
The Little Bear was introduced from Phcenicia, when the 
Pole-star became the mariner's ' cynosure.' Finally, a number 
of individual stars have Arabic appellations, dating from the 
epoch of Saracen supremacy over science. Thus * Vega,' the 
current name of the brightest star in the Greek constellation 
of the Lyre, is the remnant of an Arabic phrase signifying 
the ' Falling Eagle,' while * Altair ' stands for the * Flying 
Eagle ; ' * Deneb ' means the Tail of the Swan ; * Fomalhaut,' 
the ' Mouth of the Fish ; ' ' Eigel ' in Orion is the Foot, 

* Betelgeux,' the ' Shoulder of the Giant,' and so on. 

The constellations* now generally recognised are sixty- 
seven in number, of which forty-eight are found in Ptolemy's 
'Almagest.' From Ptolemy, too, is derived the method of 
classifying the stars by 'magnitudes.' This is a most in- 
appropriate term, since none of the stars have any perceptible 
dimensions. They are literally what Shelley calls them, 

* atoms of intensest light ' — globes shrunken by distance to 
the semblance of mere shining needle-points. Our own sun, 
removed to the place of the nearest fixed star, would be in 

» Treatise on Astronomy, p. 163, note, 

< For an easy method of identifying the chief northern stars, see Sir Robert 
Bairs Story of the Heavens, p. 372; also the * Uranography ' in Toang's 
Elements of Astronomy, 1890. 



the same condition ; its diameter of y|/' would be utterly in- 
appreciable with the largest telescope. It is true that the 
telescopic images of the stars appear to be of measurable size ; 
but this is so purely an optical effect that the ' spurious discs ' 
shown by them actually grow smaller instead of larger as / 
the power of the instrument is increased. 

Thus ' magnitude ' has nothing to do with apparent size, 
but refers entirely to apparent lustre, which depends upon 
distance and intensity of shining, as well as upon real 
dimensions. The faintest stars have the highest numerical 
magnitudes ; and it has been found that the gap between each 
successive order, as represented by the stars traditionally 
belonging to it, corresponds to a falling-oflf of light in the 
proportion of about 2^ to 1. The arrangement by magnitudes 
is, of course, entirely arbitrary ; natural gradations are not 
by a flight of steps, but along an inclined plane. Stars 
classed as of the first magnitude (of which there are ten in 
each hemisphere) * differ accordingly very much among them- 
selves. Sirius exceeds Fomalhaut no less than twelve times ; 
Vega is nearly thrice as brilliant as a Grucis. Arcturus 
outshines every other northern star, but three southern 
luminaries — Sirius, Canopus, and a Centauri— are superior to 
it. Of second magnitude are the seven stars grouped to form 
^ Charles's Wain,' the Pole-star, and the most vivid gems in 
Perseus, Cassiopeia, and the Swan. Stars of the sixth mag- 
nitude are the faintest ordinarily visible to the naked eye ; 
but those of the seventh can be seen under advantageous 
circumstances. The plan, introduced by Bayer in 1603, of 
naming the stars of the several constellations roughly in 
order of brightness by the letters of the Greek alphabet, 
established for each a kind of light-sequence, useful though 
far from exact. The smaller stars are usually distin- 


^ The ten brightest stars Dorth of the equator are : Arctarus = a Bootis, 
Vega » a T'yre, Capella » a AarigaB, Procyon =■ a Canis Minoris, Betelgeux = 
a Ononis, Aldebaran «= a Taari, Altair » a Aquilfls, PoUux a fi Oeminorum, Begu- 
lasaa Leonis, and Deneb=a Cygni. The ten brightest southern stars are: 
Sirius = a Canis Majoris, Canopus » a Argiis, a Centauri, Bigel^iS Ononis, 
Achemar « a Eridani, fi Centauri, a Crucis, Antares » a Scorpii, and Fomalhaut ^ 
a Piscis AuatrinL 5 ^Utcu r a V Li^ v^ 

B 2 


guished by tbe numbers attached to them in varioas cata- 

One of the most 'notorious eircumstances about the stars 
is their * twinkling.' They undergo, especially when near the 
horizon, extremely rapid changes of lustre, attended some- 
times by the glinting of prismatic colours. Nor do all stand, 
in this respect, on the same level. White stars twinkle more 
than red ones. Even early and untutored observers noticed 


The fiery Sirius alters hue, 
And bickers into red and emerald. 

whence it was called by Aratus ttoikiKob, the * many-coloured ; ' 
and chromatic unsteadiness was a marked peculiarity of the 
' new stars ' of 1572 and 1604. 

It is easy to see that this effect is in some way due to the 
atmosphere. Like refraction, it vanishes at the zenith; it l^ 
varies in intensity with weather and climate. The first 
rational conjecture as to its cause was made in 1667 by 
Bobert Hooke, who attributed it to irregular refractions in 
the various air-strata. More exact inquiries on the subject 
have, in recent times, led to some curious results. 
I The impressions of light on the retina last, according to 
Plateau's careful determination, 0-84 — say one-third — of a 
second. This is the limit of their individual perceptibility. 
With more frequent recurrence, they become merged indis- 
tinguishably together. But the changes visually integrated 
as scintillation succeed each other much more rapidly than 
three times in a second. Hence the need of some means of 
separating and analysing them. 

These are provided by M. Montigny's ' scintillometer,' * 
in which the principle of employing the sensibility of different 
parts of the retina for the registration of a swift succession of 
impressions is skilfully turned to account. By the rotation 
of a glass-plate obliquely inserted in front of the eye-piece of 
a refracting telescope, the image of a star viewed with it is 

* Described in Bulletin de VAciid, des Sciences, Bnixelles, t. xvii. p. 261, 
2nd ser. ; Monthly Notices, vol. xzxvii. p . 203 ; Ciel et Terre (Fievez), t. i. 
p. 869. 


made to describe a perfect circle in the field. The line of 
light thas traced out is, in the absence of scintillation, con- 
tinuous and of a uniform hue, but breaks up, under its in- 
fluence, into vividly tinted arcs, at times into prismatic 
* pearls.' The addition of a pair of crossed wires facilitates 
the reckoning of the colour-fluctuations thus rendered sepa- 
rately visible ; and they are found to occur, on an average, 
in white stars standing thirty degrees above the horizon, 
seventy-eight times in a second, in yellow and red stars 
similarly placed, sixty-eight and fifty-six times respectively.^ 

The explanation of these appearances is evidently to be 
sought in the refractive power, combined with the disturbed 
condition of our atmosphere. For a different path through 
its strata is necessarily pursued by each of the differently 
refrangible beams united to form the image of a star. The 
violet enters them higher up, since it is more bent in transit 
than the red ; and so proportionately of the rest. Each then 
is liable to encounter different vicissitudes on the way, be- 
trayed to our sight by rapid flashes of colour. Each is 
affected by innumerable small deviations and momentary 
caprices of refraction ; so that the bundle of rays picturing 
a star at a given instant is, as it were, a fortuitous and 
eminently unstable combination. It is dissolved, and a new 
one constituted, sixty or seventy times in a second; and 
the elements temporarily missing determine the resulting 

We can now understand that white stars twinkle more 
than red, because the sheaf of their beams being fuller, inter- 
ceptions of them are more frequent. But planets which are 
radiating discs, and not merely points, scarcely show the effect 
at all, because the absence of rays from one part is compen- 
sated by the arrival of rays from other parts of their surfaces. 
That stars do not scintillate in large telescopes is due to the 
neutralisation of each casual stoppage by the great number 
of the rays collected together. Instead of a twinkling image, 
however, a blurred and distended one is formed, and observa- 
tion gains nothing by the exchange. And since the degree 

* Bull, de VAcad, Bruxelles, t. xzzvii. p. 185, 2iid ser. 


to which this phenomenon is present varies very much with 
locality, the sites for powerful instraments shoald certainly 
not be chosen, as M. Exncr has remarked,' regardless of its 
prevalence. At Vienna it is particularly troublesome, and 
Dr. Pernter*s experiments go to prove that even an ascent 
to considerable altitudes affords no security against it. He 
found Sirius, in fact, actually to scintillate more at the summit 
than at the foot of the Sonnblick (10,000 feet high).« 

Scintillation, like astronomical refraction, augments as 
the thermometer falls and as the barometer rises. This is 
inevitable, since the first condition of its occurrence is differ- 
ential refractive action on the various light-rays.' But it has 
other less obviously accountable meteorological relations, 
established by M. Montigny during nearly forty years of 
observation. With the quantity of moisture in the air the 
twinkling of the stars increases so markedly as to serve for a 
useful prognostic of rain. Hurricanes and cyclones are still 
more emphatically announced by it, and it is extremely 
sensitive to magnetic disturbances.^ Ussher was struck in 
the last century, with the surprising vividness of scintilla- 
tion during auroraB ; M. Montigny finds that the coincidence 
extends to magnetic commotions perceptible only instrument- 
ally. Moreover, Weber remarked at Peckeloh in 1880, that 
stars situated near the magnetic meridian twinkled more than 
elsewhere in the sky ; * and an element of change in this re- 
spect, depending upon the points of the compass, has since 
been fully ascertained. So that it is far from improbable that 
scintillation may turn out to be one of the many terrestrial 
phenomena associated with vicissitudes in the physical con- 
dition of the sun. 

The world of stars thrown open by the telescope may 
fairly be called boundless. Using a glass only two and a half 
inches across, Argelander registered 824,189 down to 9^ mag- 
nitude, all in the northern hemisphere, enlarged by only a very 
narrow zone (one degree wide) of the southern. The work 

» Astr. Nach. No. 2,791. * Observatory, vol. xii. p. 194. 

* Montigny, BuU. de VAcad. Brnzelles, t. zlvL p. 613, 2nd series. 

* Ibid. p. 17 ; Comptes RenduSt t. xcvi. p. 673. 

* Woehenschrift fUr Astronomic, 1880, p. 294. 


was extended by Schonfeld to the southern tropic, and is now, 
nnder Dr. GilFs direction, being completed by photographic 
means to the southern pole. When this is done about 650,000 
stars will be identified, and (in Professor Holden's phrase) 
' indexed.' Each will form a possible object of future investi- 
gation, and will have been admitted (so to speak) to the 
citizenship of astronomy. 

Now Mr. Plummer has shown ' that the lucid stars (or 
those visible to the naked eye) in Argelander's ' Durch- 
musterung ' give as much light as 7,S49, the telescopic stars aR 
23,387 sixth magnitude stars. So that those singly invisible 
really illuminate the sky just three times as much as those we 
can see and distinguish. Summing up the entire light of this 
grand collection of S24,0(X) stars, we get for its equivalent ^^r 
of the light of the full moon,' and for the total from all the 
650,000 stars to 9^ magnitude, -^ full moonlight. 

What amount of scattered effulgence reaches us from still 
fainter stars, there is at present no sure means of estimating ; 
yet in putting the inclusive aggregate at one-tenth full moon- 
light M. Gustave I'Hermite^ almost certainly overshot the mark. 
The number of such objects is entirely unknown. All that 
can be stated with any confidence is that there are many 
millions of them. The largest telescopes in the world, it is 
thought, might show sixty million stars over the entire sky ; 
but the margin of doubt is very wide. M. THermite has, 
however, computed the population of the stellar universe 
from his valuation of stellar light-power, and finds it, on the 
assumption that the scattering of the stars is everywhere 
just such as it is in our own neighbourhood, to be sixty-six 
thousand millions ! This extravagant result indicates, with- 
out doubt, that both valuation and assumption are erroneous, 
although the inquiries based upon them are interesting and 
ingenious. Especially noteworthy is the fact that each class 

' Monthly Notices, vol. xxxvii. p. 436. 

' Mr. Plummer's result appears to be vitiated by his adoption of too low a 
value for the light of Sinus. That given in the text is derived from combining 
Pickering's photometric determination of Vega with SeidePs measures of the 
same object, showing it to be equal to ^l^ full moonlight. 

• L'Astroncmiej t. v. p. 406. 


of stars sends us appreciably more light than the class next 
above it. The light aggregate of second magnitude stars 
exceeds that of first, of third that of second, and so on. The 
fainter the stars, in short, the greater is their total luminous 
power,' because their augmented numbers more than counter- 
balance their diminished individual lustre. But this pro- 
gression, it is evident, cannot go on indefinitely, since 
otherwise an indefinitely intense radiance would fill the sky. 
Darkness would be abolished through the shining of invisible 
stars. It follows that the observed order of the stellar world 
has assignable limits — that the star-depths, however profound, 
are not absolutely unfathomable. 

The task of exploring them is not then altogether hopeless. 
It can never indeed be exhausted ; but it can fairly be grappled 
with by finite minds. It does not evade their efforts with the 
passive scorn of material infinitude. Genuine, if partial, suc- 
cesses have crowned them in the past, and will, it may be 
hoped, continue to crown them in the future. 

We must not, however, in seeking encouragement from the 
thought that it does not utterly defy our powers, underrate 
the arduousness of the enterprise we have engaged upon. The 
nature of our own sun offers a vast and intricate problem, 
still very fax from being solved ; but stellar space contains 
many millions of suns, variously constituted, variously circum- 
stanced, exceeding for the most part perhaps very greatly in 
size and splendour the body we depend upon for vital neces- 
sities. Now each of these millions of suns challenges the 
closest personal attention ; no single one of them is exactly like 
any other, and their differences and resemblances open end- 
less vistas of instruction and interest. Their incredible remote- 
ness in no way derogates from their real dignity. An all but 
evanescent speck of light in the field of the great Lick refractor 
may be the life-giving centre of a system of worlds, each 
abounding as marvellously with proofs of creative wisdom and 
goodness as the far-away little planet in which our temporal 
destinies are rooted. Such light-specks are then equally 
deserving of study with the most effulgent orbs in the sky, 

' L^ Astronomie, t. v. p. 409. 


although it may never be practicable to bestow it upon them. 
We can scarcely indeed imagine the amount of telescopic im- 
provement which would be needed in order to bring them 
within the range of critical examination. For the present, at 
any rate, physical research must be confined to some thousands 
of the brighter stars which may serve as specimens of the rest. 
Nor need we lament the restriction. Generations of workers 
might expend their energies in gathering facts from the field 
actually open to them, and yet leave a full harvest for their 
successors. In all experimental inquiries, it may indeed be 
said that the reaper, as he garners one crop of knowledge, 
BOWS another : so endless are the secrets of nature, so untiring 
the inquisitiveness of man. 

The stars in their combinations demand inquiry no less 
than the stars in themselves. Stellar systems are to be met 
with in indescribable profusion and variety, from mutually 
circling pairs, through groups including thousands of physi- 
cally related objects, to the stupendous coUection of groups 
which we call the Milky Way. But as yet investigation has 
barely skirted the edge of this well-nigh infinite region. Before 
it can be penetrated by so much as a plausible conjecture, 
statistics are wanted of the distances and movements of thou- 
sands, nay millions of stars. 

Nor is the amassing of them any longer the desperate enter- 
prise it seemed a short time since. By the unhoped-for develop- 
ment of novel methods, the pace of inquiry has been quickened 
all along the line. Particulars are accumulated faster than 
they can be assorted and arranged. Time seems to have 
multiplied itself for the purpose of gratifying curiosity which 
becomes keener as its sublime objects loom more distinctly above 
the horizon of thought. Ten years now count for a century 
of the old plodding advance. Express trains carry passengers 
on errands of research, as well as of business or pleasure. 
Problems ripen as if in a forcing-house, and so numerously as 
almost to bewilder the attention. 

The whole subject of sidereal natural history is indeed wide 
and intricate beyond what it is easy to convey to those ap- 
proaching it for the first time. There is scarcely a topic in 


physical astronomy with which it is nnconnected. The pro- 
gress of discovery has gradually drawn closer the generic 
relationships of the heavenly bodies. The sun has come to be 
recognised as the grand exemplar of the stars; meteorites 
show themselves to be intimately associated with comets; 
comets incur a ' vehement suspicion ' of kinship with nebulae; 
while the stellar and nebular realms blend one with the other 
as indistinguishably as the animal and vegetable kingdoms of 
organic nature. 

The strange cloud -like objects called ' nebulae ' may be con- 
sidered as wholly of telescot t-- revelation. Only one of them 
— the famous object in the girdle of Andromeda — can be at all 
easily seen with the naked eye; and even that escaped the notice 
of all the Greek, and most of the mediaeval astronomers. The 
' nebuloeae ' of the ancients were many of them small groups 
of stars accidentally set close together ; but among the seven 
enumerated by Ptolemy were two real clusters like the 
Pleiades, only (presumably) much farther away, one in Perseus, 
and the other in Cancer. Indeed, to extremely short-sighted 
persons, the Pleiades themselves put on a nebulous appearance, 
the individual stars running together into one wide blot of 

Halley was the first to form anything like an adequate 
conception of the importance of nebular observations. He 
was acquainted in 1716 with six ' luminous spots or patches, 
which discover themselves only by the telescope, and appear 
to the naked eye like small fixed stars; but in reality are 
nothing else but the light coming from an extraordinary great 
space in the ether, through which a lucid medium is diffused, 
that shines with its own proper lustre.' ' Only two of Halley's 
half-dozen objects, however— those in Orion and Andromeda — 
were genuine nebulae ; the rest when viewed with better instru- 
ments than his * six-foot tube,' proved to be magnificent star 

This small beginning of knowledge was followed up by 
Lacaille in the southern, by Messier in the northern hemi- 
sphere. But Herschel's great telescopes opened the modern 

» Phil, Trans, vol. xxix. p. 390. 


epoch in the science of nebuIaB. They began as the result 
of his labours to be counted by the thousand instead of by 
the score. Portions of the sky were found to be crowded 
with them. Yet the vast majority must always, owing to their 
extreme faintness, remain imperceptible without powerful 
optical aid, only sixty-four coming into view with the same 
telescope which showed Argelander 324,000 stars. Thus 
access to the nebular heavens is granted conditionally upon 
making the very most, by the employment of large telescopes, 
of the little light they send us. No wonder, then, that com- 
paratively little progress has been made with their investiga- 
tion. Upwards of eight thousand nebulae, it is true, have been 
discovered, but for the most part, only to be again neglected. 
Assiduous and prolonged observation is, nevertheless, indis- 
pensable for the detection of the cyclical or progressive changes 
doubtless proceeding in these inchoate systems. Nor have 
mere visual records proved sufficient for the satisfactory treat- 
ment of so delicate a subject. Other methods, at once more 
stringent and more expeditious, were needed, and have now 
happily been made available. With their help, it is already 
evident that knowledge will advance by rapid strides. 

This then is the task of sidereal astronomy — to investigate 
the nature, origin, and relationships of sixty million stars 
and upwards of eight thousand nebulae — to inquire into their 
movements among themselves, and that of our sun among 
them — to assign to each its place and rank in the universal 
order, and, gathering hints of what has been and what will 
be from what is, distinguish hierarchies of celestial systems, 
and thus at last rise to the higher synthesis embracing the 
grand mechanism of the entire — the sublime idea of Omnipo- 
tence, to which the stars conform their courses, while * they 
shine forth with joy to Him that made them.' 




Sidereal science is, on its geometrical side, of modern develop- 
ment ; on its physical side, of modem origin. The places of 
the stars, as referred to certain lines and points on the surface 
of an imaginary hollow sphere, are obtained now on essentially 
the same principles as by Hipparchns, only with incomparably 
greater refinement. And refinement is everything where the 
stars are concerned. Significant changes among them can only 
be brought out by minute accuracy. To a rough discernment 
their relative situations are immutable ; and systematic 
inquiries into their movements hence became possible only 
when the grosser errors were banished from observation. 
Bessel's discovery of Bradley's exactitude gave the signal for 
such inquiries. It became worth while to re-observe stars 
already so well determined that discrepancies might safely be 
interpreted to mean real change. 

Thus it is only within the last sixty or seventy years that 
the stars have been extensively catalogued for their own sakes, 
and no longer in the undivided interests of planetary or 
cometary astronomy. The scope of such labours now widens 
continually. For the objects of them are all but innumerable, 
and the nineteenth century has brought to bear on its large 
schemes of scientific ambition heretofore undreamt-of facilities 
for executing them by combination. The project set on foot 
by the German Astronomical Society in 1867 of fixing the 
precise places of all stars to the ninth magnitude, has found 
co-operators in all parts of the world ; when it is completed 
so far as to join on to the southern * zones,' observed by Gould 
and Gilliss, not far from 400,000 stars will be not merely 
recorded, but known in the strict sense of modern astronomy. 


Even this is not enough. There is a prospect that, early in the 
coming century, stars, approximately placed, will be counted 
by millions, while further unidentified milhons still congregate 
unheeded in the unexplored recesses of the sky. 

A star is located in the heavens, just as a city or a moun- 
tain is located on the earth, by measurements along two 
imaginary circles. Its 'declination,' or distance from the 
celestial equator, corresponds to terrestrial latitude ; its ' right 
ascension ' to terrestrial longitude. The astronomical prime 
meridian passes through the first point of Aries, that is, through 
the sun*s position at the vernal equinox ; intervals from it are 
reckoned eastward from 0° to 860°, or in time from 0** to 24**. 
And since the zero-point retreats slowly westward by the effect 
of ' precession,' it follows that the right ascensions of nearly 
all stars increase steadily year by year, apart from any move- 
ments ' proper ' to themselves. 

The diurnal revolution of the sphere furnishes the sole 
standard of time in sidereal astronomy. Sidereal noon at 
each spot on the earth is the moment when the first point of 
Aries crosses the meridian of that spot ; the right ascensions 
of the heavenly bodies indicating the order of their successive 
culminations. Thus, if the right ascension of a star be two 
hours and twelve minutes, it will always cross the meridian 
of any given place two hours and twelve minutes after the first 
point of Aries has crossed it, coming up behind it, to that extent, 
in the grand diurnal procession. Differences in right ascen- 
sion signify accordingly differences in times of culmination ; 
and their measures in hours, minutes, and seconds, need only 
multiplication by fifteen (the number of times that 24 is con- 
tained in 860), to appear as measures of arc in degrees, 
minutes, and seconds. 

A transit-circle and a clock are the two essential instru- 
ments for determining the places of the stars. The instant, 
to the tenth of a second, at which a star stands in the 
meridian, is noted ; the vertical circle is read, showing its 
* zenith-distance ' (giving at once its declination when the 
latitude of the observatory is known), and the observational 
part of the work is done. The data thus obtained, after 


undergoing numerons corrections, suffice to determine the 
position of the star with reference to some other * fundamental ' 
star, the ai>solute place of which has been separately and 
laboriously ascertained. 

This business of star-location forms the substratum of the 
older astronomy. But the precision given to it is altogether 
new, and alone has fitted it to be the means of eliciting facts 
80 coy as those that relate to stellar movements. For the 
disclosure of them subtle devices of accuracy are needed 
which our astronomical progenitors never cast a thought upon. 
Optical and mechanical skill have, in our days, reached a 
point of almost ideal perfection ; yet when the artist has done 
his utmost, the instrument is, in a sense, still in the rough. 
The astronomer then takes it in hand, and his part is often 
the more arduous and anxious. The investigation of small 
surviving errors, the contrivance of methods for neutralising 
their effects, the carrying out of delicate operations of adjust- 
ment, the detection of microscopic deformations, tremors of 
the soil, inequalities of expansion by heat, fall to his share. 
Even his own rate of sense-transmission has to be deter- 
mined, and figures, under the title of ' personal equation/ as 
a correction in the final result. For, between the actual 
occurrence and the perception of a phenomenon, there is 
always a gap, more or less wide, according to individual 
idiosyncrasy, and it is only after this gap— tiny though it be 
— has been crossed, that electricity can be caUed upon to play 
its prompt part as amanuensis to the observer. 

This detailed and painful struggle against error has alone 
made sidereal astronomy possible, by precipitating from the 
mixed solution that held them the minute quantities it deals 
with. Just because the universe is almost infinitely large, 
these quantities are almost infinitely small. They are small, 
not in themselves, but through the incomprehensible remote- 
ness of the bodies they affect. 

Geometrical sidereal astronomy deals with the motions 
of the stars. But these are of different kinds. 'Proper' 
motions — so-called to distinguish them from 'common' 
apparent displacements due to the slow shifting of the points 


of reference on the sphere— advance uniformly along a great 
circle; orbital revolutions of one star round another are 
periodical in small ellipses ; besides which annual oscillations, 
varying in extent with the distances from ourselves of the 
objects performing them, are barely measurable in a few of 
the nearest stars. For determinations of these orbital and 
' parallactic ' movements such microscopic accuracy is indis- 
pensable as has only recently been attained. The instruments 
employed in them are the equatoreal with micrometer attached, 
and the heliometer. 

An equatoreal is a telescope so mounted as to follow the 
diurnal revolution of the heavens. It is connected with an 
axis directed towards the pole, and revolving by clockwork 
once in twenty-four hours. So that an object once brought 
into the field of view remains there immovably for any desired 
time, provided that the telescope be clamped in position, and 
the clock set going. The inconvenience of the earth's rotation 
in producing a continual ' march-past ' of the heavenly bodies 
is thus neutralised. 

To the eyeend of an equatoreal is usually attached an 
arrangement of spider lines constituting a ' filar micrometer.' 
Two sets of such threads (which in subtlety and evenness of 
texture far surpass any artificial production), crossing at right 
angles and some of them movable by fine screws, while the 
whole can be made to revolve together, afford a most delicate 
means of ascertaining the distance and direction from each 
other of any two objects close enough for simultaneous observa- 
tion. Measures of double stars are executed, and some stellar 
parallaxes have been determined in this way. But for the 
latter purpose, the * heliometer * is the more appropriate 

Its designation is a misnomer, or rather represents the 
tradition of an original purpose to which it was never effec- 
tively applied. The true function of a heliometer is the critical 
measurement of two adjacent stars, or of a star and planet. 
Primarily, it is an equatoreal telescope; its micrometrical 
powers are conferred by the division of the object-glass into 
two halves sliding along their common diameter, and dupli- 


eating by their separation the combined image formed by them 
when together. The amount of movement given to the seg- 
ments in bringing about alternate coincidences between oppo- 
site members of the pair of stars shown by each, suffices to 
determine with the utmost nicety the interval between them. 
That is to say, after endless precautions for accuracy have been 
taken, and endless care bestowed upon detecting and obviating 
occasions of infinitesimal error. 

The Radcliffe Observatory at Oxford possesses the largest 
heliometer in existence. The diameter of its object-glass is 
seven and a half inches. A similar instrument, however, erected 
at the Boyal Observatory, Cape of Good Hope, nearly forty 
years later, is but slightly inferior in size, and is in other re- 
spects considerably its superior. Dr. Elkin, at Yale College, 
has charge of the only heliometer in the New World, while 
several are to be found in Germany and Russia. The 
Bepsolds of Hamburg may be said to hold a monopoly in the 
mechanical part of their production ; and Merz of Munich 
stands almost alone among opticians in his readiness to 
take the responsibility of sawing a fine object-glass in two. 
Nor is the aptitude for the use of these instruments by 
any means universal among observers; hence their com- 
parative scarcity. 

The science of the motions of the stars is only a part of 
modern sidereal astronomy. Within the last quarter of a 
century, a science of their nature has sprung up and assumed 
surprising proportions ; a science the reality of which confounds 
forecast, yet compels belief. Sidereal physics has a great 
future in store for it. Its expansiveness in all directions is 
positively bewildering. The * What next ? * is hardly asked, when 
it is answered often in the least looked-for manner. In follow- 
ing its progress, the mind becomes so inured to novelties, that 
antecedent improbability ceases to suggest dissent. Some 
details of what we have thus so far learnt will be contained 
in the ensuing chapters ; the means employed must be briefly 
indicated in this. 

They are of three principal kinds — spectroscopic, photo- 
metric, and photographic. The general theory of spectrum 


analysis has been explained elsewhere;* here we need only re- 
peat that it rests apon the constant position in the spectrum 
of the rays of light given out, when in a state of vaporous 
incandescence, by each separate chemical substance. These 
invariable lines serve as an index to the presence of the sub- 
stance they are associated with, in the sun or in a star, no less 
than in the laboratory. Whether they be bright or dark, the 
principle remains the same. They are bright when the ignited 
vapour originating them is the chief source of illumination ; 
dark, when a stronger light coming from behind is absorbed by 
its interposition. Their appearance as ' lines ' is merely due 
to the transmission through a narrow slit of the Ught after- 
wards prismatically dispersed. 

Now a main difficulty in getting starlight to disclose its 
secrets, is that there is so Uttle of it. It will not bear the 
necessary amount of spreading-out, but evades analysis by 
fading into imperceptibiUty, like a runnel of water that widens 
only to disappear. Hence the absolute necessity in stellar 
spectroscopy for large telescopes. The collecting nets have to 
be widely extended to gather in a commodity so scarce. 
Gould we at all realise, indeed, the portentous expanse of the 
ever-broadening sphere filled by the stellar beams as they 
travel towards us, we should be inclined to wonder, not at 
their faintness, but at their intensity. But the weakening 
effect of distance is in some degree counteracted by powerful 
concentration ; and this is one of the chief uses of the large 
telescopic apertures so much in vogue at the present time. 

Viewed with the Lick refractor of thirty-six inches, any 
given star is 82,400 times brighter than it appears to the naked 
eye, or 824 times brighter than when shown by a two-inch 
telescope.* The large instrument, that is to say, provides 824 
times as much material for experimenting upon as the small, 
or places upon the same level of advantage for purposes of 
scrutiny objects 824 times as faint. 

The interpretation of spectral hieroglyphics, by which we 
learn the chemical constitution of a star, is a very delicate and 

* See the author's Popular History of Astronomy, 2nd edit. p. 175. 
' Holden, English Mechanic, vol. xlvi. p. 628. 



laborious operation. What is called a ' comparison-spectrum ' 
is asually employed as an adjonct to it. Rays from some 
terrestrial source are reflected into one half of the slit, through 
the other half of which the stellar rays are admitted. Both sets 
then traverse the same prisms, and form strictly comparable 
spectra side by side in the same field of view. Lines common 
to both can thus easily be identified ; and their genuine occur- 
rence leaves no doubt that the element compared —hydrogen, 
sodium, iron, magnesium, or any other — enters into the compo- 
sition of the star. But this process of matching can seldom or 
never be completely carried out. A dozen known lines may be 
attended by a hundred unknown ones, either too faint to be 
distinctly seen, or in positions unfamiliar to terrestrial light- 
chemistry. Nor is it safe to infer the absence of an ingredient 
from the absence of its representative rays. Many causes may 
contribute to render the display of lines in steUenspectr a selective. 
Where direct comparisons can be dispensed with, a slit is 
not essential to stellar light-analysis. For a star, having no 
sensible dimensions, gives rise to none of the confused over- 
laying of images produced by grosser light-sources imless 
• superfluous rays be excluded by the use of a fine linear aper- 
ture. Hence the possibility of applying a ' slitless spectro- 
scope ' to the stars. Their light is then simply passed through 
a prism, either before it enters, or as it leaves the telescope. 
The resulting variegated stripe, looked at through a cylindrical 
lens to give it some tangible breadth, shows the dark gaps or 
lines significant of the ' type ' of the star. 

But prismatic analysis is not merely communicative as to 
the physical and chemical nature of the stars. It can tell 
something of their movements as well. And, what is especi- 
ally fortunate, the information that it gives is of a kind other- 
wise inaccessible. 'End-on' motions, as every one knows, 
are visually imperceptible ; the discovery that the spectro- 
scope has the power to make them sensible is of such far-reach- 
ing importance that Dr. Huggins, by bringing the method 
into effective operation, performed perhaps the greatest of his 
many services to science. Through the link thus established, 
geometrical and physical astronomy have been placed in 


closer mutual relations than could have been thought possible 

The observations concerned are of great delicacy, and can 
only be made with a powerful telescope, collecting light suffi- 
cient to bear a considerable amount of dispersion. Their 
object is to measure the minute displacements of known lines 
due to ' radial ' or end-on motion, and proportional in amount 
to its velocity. These displacements are towards the blue end 
of the spectrum when the star is approaching, towards the 
red when it is receding from the earth. The refrangibility of 
the luminous beams is changed, in the one case, by the 
crowding together of the ethereal vibrations, rendering them 
more numerous in a given time, in the other, by their being 
(as it were) drawn asunder, and so rendered less numerous. 
The juxtaposition of a standard terrestrial spectrum, such as 
that of hydrogen, gives the means of measuring deviations 
thus produced, and so of determining the rate of approach or 
recession of the star examined. But the process is impeded 
to a degree hardly imaginable without personal experience 
by troubles in the ocean of our air. The twinkling of the 
stars is represented in their spectra by tremors and undula- 
tions often permitting only instantaneous estimates of line- 
positions. But for this inconvenience an unexpected remedy, 
as we shall presently see, has been found. 

Stellar photometry has a two-fold object. It gives the 
means of investigating, first, the individual nature ; secondly, 
the collective relations of the stars. Stellar lustre is affected 
by endless gradations of change. It is rarely, perhaps never, 
really constant. Periodical fluctuations are in many cases 
obvious ; secular variations are suspected. The suspicion can 
be verified only by precise light-measurements repeated at 
long intervals. 

Their application to the problems of stellar distribution 
becomes feasible through the dependence of brightness upon 
distance. The law of the decrease of light with the increase 
of the square of the distance, is universally familiar. If all 
the stars were equal in themselves, their apparent differences 
would thus at once disclose their relative remoteness. We 

c 2 


could locate them in space just as accurately as we could 
determine their lustre. But in point of fact the stars are 
enormously unequal, and we can hence reason from distance 
to brightness only by wide averages. A statistical method 
alone is available, and its employment involves the establish- 
ment of strict principles of light measurement. 

The first requisite for this purpose was an unvarying and 
consistent scale, provided with the least possible disturbance 
to existing habits of thought, by regularising the antique 
mode of estimation by ' magnitudes.' Intervals loosely defined 
and unequal were made precise. A ' light-ratio ' was agreed 
upon. To this proportion of change from one magnitude to 
the next, the numerical value 2*612 ' has been assigned. 
That is to say, an average first magnitude star sends us 2*512 
times as much light as an average star of the second magni- 
tude, which, in its turn, is 2*512 times brighter than one of the 
third, and so on. From the first to the third magnitude, the 
step is evidently measured by the square of the * light-ratio ' 
(2*512 X 2*512 =6*310) ; and in general the relative brilliancy 
of any two stars may be found by raising 2*512 to a power 
represented by the numerical difference of their magnitudes. 
One first magnitude star, for instance, is equivalent to one 
hundred of the sixth rank ((2*512)^ = 100); and to no less 
than a million stars of the sixteenth magnitude. 

All this is a matter of pure definition, and definition is a 
useful leading-string to experiment. It is something to have 
a clear conception in the abstract, of what a tenth, eleventh, 
twentieth magnitude star is, even though the conception be 
not altogether easy to realise. The problem of applying the 
numerical standard set up was practically solved almost at 
the same time by Professor Pritchard at Oxford, and by Pro- 
fessor Pickering at Cambridge in the United States. They 
first systematically and extensively employed instrumental 
means in stellar photometry, with the result of satisfactorily 
ascertaining the comparative lustre of all stars visible to the 
naked eye in these latitudes. 

Professor Pritchard adopted for his researches the * method 

' Selected as the namber of which 0'4 is the logarithm. 


of extinctions.' The image of each star was made to vanish 
by sliding between it and the eye a wedge of neutral-tinted 
glass, the thickness of which needed to produce extinction was 
found to give a very exact measure of intensity. In this way 
the brightness of 2,784 stars from the pole to ten degrees 
south of the equator was determined and registered in the 
* Uranometria Nova Oxoniensis.' 

The Harvard 'meridian-photometer' (a modification of 
Zollner's) was constructed on the principle of * equalisation.' 
The images of the pole-star (adopted as a standard of com- 
parison) and of each star successively experimented upon, 
were reflected into a fixed telescope, and brought to an exact 
equality by means of a polarising apparatus. From the 
amount of rotation given, for this purpose, to the double- 
refracting prism, the difference of original light-power was 
easily deduced. The method is of wider applicability than 
that by extinctions ; none the less, the * wedge photometer,' 
in the form given to it by Professor Pritchard, has taken its 
place as an indispensiable adjunct to such inquiries. With 
either instrument the limit of clearly distinguishable difference 
is about one-tenth of a magnitude. 

The Harvard photometry * includes all stars to the sixth 
magnitude as far as 80"" of south latitude, to the number of 
4,260. But the more southerly among them were evidently 
observed at a great disadvantage, owing to their low altitude, 
and the consequent heavy light-tax exacted from them by 
the atmosphere. A process of * reduction ' was then needed 
before the stars in different situations could be fairly com- 
pared. The recorded brightness of each was 'reduced,' 
according to a scale of correction laboriously constructed from 
experiment, to what it would be in the zenith. Even here 
the loss of light is about twenty per cent., but it increases to 
thirty-six at 30°, to sixty per cent, at 13^° above the horizon.* 

In the Harvard photometry only two stars, Aldebaran 
and Altair, are rated as strictly of the first magnitude. Each 

^ Harvard Annals, vol. xiv. pt. i. (1884). For a comparison with the Oxford 
results see ibid, vol. xiii. p. 15. 

* Pritchard, Memoirs R. Astr. Society, vol. xlyii. p. 372. 


of them gives just three timee as much light as the pole- 
star. They have in the northern hemisphere five superiors — 
Arcturus, Gapella, Vega, Procyon, and Betelgeux — the standing 
of which has accordingly to be expressed by fractional numbers. 
Gapella and Vega are of magnitude 0*2, signifying that each 
is eight-tenths of a magnitude brighter than Aldebaran or 
Altair, while the O'O attached to Arcturus means that it 
excels either of these stars by one whole magnitude. Carrying 
out the same system of notation, we get negative numbers for 
the designation of still higher grades of lustre. The fact, for 
instance, that we receive from Sirius more than nine times 
as much light as from Aldebaran, corresponding to a superiority 
of two magnitudes and four-tenths, is compactly expressed 
by calling its magnitude —1-4. To find a star outshining 
Sirius, we must go to our own sun, to which a rank can be 
assigned on the same scale. Its light, as measured by Alvan 
Clark in 1868, exceeds that of the dog-star 8,600 million 
times. Bond made the disproportion 6,970, Steinheil 8,840 
millions to one. From a mean of the three, Professor 
Pickering fixed the sun's stellar magnitude at — 25*4,' its 
superiority over Sirius amounting to 24, over Arcturus to 
26*4, grades of ascent. 

A third photometric review of the northern heavens was 
begun at Potsdam in 1886 by Drs. Miiller and Eempf.' 
Since stars down to 7*6 magnitude are embraced by it, it 
will considerably overlap the two earlier works of the same 
kind, and will include a re*determination of all their stars, 
uncertainty as to the light-rank of which will thus be reduced 
to a minimum. Zollner's polarising photometer is the in- 
strument employed, perhaps for the last time in such a 
comprehensive research — not from any intrinsic incapacity 
for even more extended usefulness, but because better means 
are at hand. 

The invention of the telescope itself does not mark an 
epoch more distinctly than the admission of the camera into 

* Proceeddngs American Academy, vol. xvi. p. 2. 

^ Vierteljahrsschri/t Astr, QeselUchaft, vols. xxii. p. 145, xxiii. p. 124, 
zxiy. p. 141. 


the celestial armoury. All the conditions of sidereal research , 
in especial, are being rapidly transformed by its co-operation. 
The versatility of its powers is extraordinary ; no task has 
yet found it unready or incapable. It is the very Ariel of the 
astronomical Prospero. 

This untiring serviceableness was made possible by the 
substitution in 1871 of gelatine for collodion as the vehicle 
for the salts of silver, the decomposition of which under the 
influence of light forms the essential part of the photographic 
process. The new plates were, however, first used for ' astro- 
graphical ' purposes by Dr. Huggins in 1876. Since they are 
five times more sensitive dry than wet, exposures with them 
can be indefinitely prolonged. They may, besides, be prepared 
any desirable time before, and developed any desirable time 
after exposure, thus accommodating themselves in a really 
wonderful way to the needs of astronomers^ 

The unique power of the photographic plate as an engine 
of discovery is derived from its unlimited faculty for amassing 
faint impressions of light. By looking long enough it can 
see anything there is to be seen. Captain Abney's experi- 
ments convince him that no rays are too feeble to overthrow 
the delicate molecular balance of silver bromide if only 
their separately evanescent effects get sufficiently piled-up 
through repetition.^ By this power of accumulation the 
camera leaves the eye far behind. With any given telescope 
much more can be photographed than can be seen, and the 
opening of a region of research, beyond the range of visual 
accessibility, appears to be close at hand. Already the de- 
picting of the invisible has, in some isolated cases, been 
realised ; it will perhaps before long be pursued by methods 
and with results peculiar to itself. 

The penetration of space has, indeed, even thus limits. 
It is unlikely, for instance, that external galaxies (if such 
there be) will ever reveal themselves on our plates. A ne pins 
ultra is imposed, if no otherwise, by the restricted possibilities 
of continuous exposure to the sky. Darkness does not last 
indefinitely; the time during which a given object (unless 

* Observatory, vol. xii. p. 166. 


situated very near the pole) is high enough above the horizon 
for portrait-taking purposes is still shorter. Mr. Roberts has, 
however, adopted in certain cases and with a measure of suc- 
cess, the expedient of successive exposures given to the same 
plates on different nights. 

The telescope forming the image which imprints itself upon 
the prepared plate, is always equatoreally mounted, and has 
a motion given to it exactly concurrent with the revolution 
of the sphere. Yet the utmost mechanical ingenuity cannot 
make the concurrence absolutely perfect. Minute inequalities 
survive and need intelligent correction. Even more sensible 
are disturbances caused by the changes of atmospheric re- 
fraction with the ascent towards, or decline from the meridian 
of the objects in course of delineation. For these reasons a 
photographic telescope has, as a rule, a guiding telescope 
attached to its axis, through which an observer watches to 
counteract, almost to anticipate, nascent tendencies to dis- 
placement. The strain upon the attention is severe; its 
endurance, upon occasions, for three, even four hours at a 
stretch, is no small proof of resolution. The effect, however, 
of long exposures can be obtained in shorter times by giving 
increased sensitiveness to the plates, which have now been 
brought to such a pitch of * rapidity ' that their * fogging ' from 
the general illumination of the sky, or the glare of a great 
city, is a danger that has to be carefully guarded against. 

Exposures can also be curtailed by shortening the focus 
of the photographic telescope, the image being thus rendered 
smaller, and — through the closer concentration of the same 
amount of light — more intense. For simply exploring the 
skies, sounding their depths, and dredging-up their contents, 
nothing can be better than the form of an ordinary portrait-lens. 
With such a one, only two inches in diameter, the picture of 
the comet of 1882 was taken at the Royal Observatory, Cape 
of Good Hope, the ' thick-inlaid ' background of which was 
the first palpable revelation of the star-charting powers of the 
camera ; and much of Professor Pickering's admirable work 
in sidereal photography has been done with a Voigtlander's 
' doublet ' (two achromatic lenses in combination) of eight 


inchea aperture and about forty-five focus. With this instru- 
ment, objects imperceptible through the Harvard fifteeu-inch 
refractor are photographed; a thousand stars within one 
degree of the pole have been catalogued where some forty 
were previously located ; ' and to eighteen known nebulaB 
twelve till then unknown were added within a region about ^h^ 
of the area of the heavens. In the ' Bruce telescope/ about 
to be erected under Professor Pickering's direction on one of 
the mountains of Southern Galifomia, under almost ideal atmo- 
spheric conditions, this plan of construction will be carried 
out on a larger scale. The object-glass will be twenty-four 
inches in diameter, the focal length eleven feet. Stars fainter J 
than have ever before been reached, visually or photo- 
graphically, are expected to make their appearance on plates 
exposed to the powerful concentration of light thus effected. 
The portrait-form of lens has the additional merit of giving a 
large field of view. Each photograph taken with the Bruce 
telescope will cover five degrees square (25 square degrees) on 
a scale of one minute of arc to a millimetre. The whole 
heavens could be charted on about two thousand such plates. 
Where accurate measurements are aimed at, however, the 
type of instrument represented by the MM. Henry's photo- 
graphic telescope is preferable. The object-glass in this is of 
the ordinary achromatic kind, but corrected with reference to 
chemical, instead of to visual action. The rays selected to be 
brought to a focus are those to which, not the human, but the 
* photographic retina ' is sensitive. The aperture is thirteen 
inches, the focal length eleven feet ; a plate-holder is substituted 
for an eye-piece, while a guiding-telescope of slightly inferior 
dimensions is enclosed within the same rectangular tube. 
The field of view with the Paris photographic telescope within 
which definition may be considered as virtually perfect, is a 
circle three degrees in diameter,^ covering an area of not quite 
five square degrees. Fully ten thousand of these plates would 
be needed to picture the sphere. 

* Mr. BobertB has charted 1,270 stars oyer the same space from a negative 
taken at MaghuU, August 14, 1888. Monthly Jfotices, vol. zlix. p. 10 

* BtUletin Astronomiqtie, t. vi. p. 303. 


A score of such inBtruments will presently be employed in 
all parts of the world, in carrying out the international star- 
charting operations set on foot by the Paris Congress of 1887. 
Their completion, in four years or less, will place at the dis- 
posal of astronomers two distinct series of plates, each in 
duplicate to obviate errors. The longer-exposed series, de- 
scending to the fourteenth magnitude, and including, it is 
anticipated, twenty-five million stars, may be invoked 
centuries hence for the decision of cases of suspected change. 
Were a register of the kind dating from Tycho's or Kepler's 
time now extant, it is scarcely too much to say that the whole 
fabric of stellar astronomy would be reared upon comparisons 
and consultations with it. Realising, then, the value of what 
it has power to execute, the present generation owes it to 
posterity to lay up for their use a store of facts that cannot 
fail to ripen into knowledge. 

The second series of international plates will provide the 
materials for a great catalogue of about a million and a half 
of stars to the eleventh magnitude. In taking, no less than 
in measuring them, precision will be safeguarded with the 
most anxious care. Among the special means for securing 
it, M. 0. Lohse's device of the * reticle ' deserves particular 
mention. Each plate will carry a latent image of a system of 
lines at known distances (about five millimetres apart) which, 
developed with the imprinted stars, will afford a secure index 
to possible slight distortions of the film. Such distortions are 
the bete noire of the celestial photographer ; but once made 
manifest, they become innocuous. 

The situations in the sky of the million and a half stars 
thus depicted can be ascertained by micrometrical measures 
of their positions with reference to certain known stars suitably 
distributed on the plates. The accuracy attainable in this 
way is sufficiently described by saying that it fully meets the 
requirements of modem research. These million and a half 
of new recruits will then be definitely entered on the muster 
roll of the celestial array. Each of them will have a ' local 
habitation and a name '— or at least a number. They will be 
individualised and pursued from century to century by curious 


inquiry. Their changes can no longer lurk unheeded, and 
the record of their existence, outlasting the destruction or 
decay of the negatives from which it has been derived, may 
be said to be for all time. 

Should Professor Pickering's plan of work with the Bruce 
telescope be executed, a third set of plates, showing, it is 
expected, all stars to the seventeenth magnitude north of thirty 
degrees south declination, will be added to the astronomer's 
archives. Their comparison with the international long- 
exposure plates ought to prove of the highest interest. Stars 
of seventeenth are, by definition, nearly sixteen times fainter 
than stars of fourteenth magnitude, and they should be, were 
there no failure of the star-supplies, sixty-three times as 
numerous. So that, if the international plates contain twenty- 
five millions, the Bruce plates ought (if they really embrace 
seventeenth magnitude stars) to contain twelve hundred mil- 
lions for three-quarters of the sky. There are, indeed, strong 
indications that no such portentous multitudes will make their 
appearance ; but it wQl be an immense point gained to have 
made sure that they are not there. 

* Studies of the distribution of the stars,' Professor Picker- 
ing remarks, ' can now scarcely be undertaken in any way 
except by photography.* But photography, to be really in- 
structive on this point, must be combined with photometry. 
The portrayal of millions of stars projected side by side on a 
spherical surface tells us little or nothing of their relations to 
the immensity of space. This can only be found out, for the 
vast majority of them, by collecting statistics of the amount 
of light they send us. Hence the importance of the photo- 
metry of small stars. Yet no visual means have hitherto 
proved competent to deal with it. Eye-estimates, however 
guided and succoured by instruments, break down when 
pushed too far down the scale. The problem is evidently one of 
those reserved for successful treatment with the camera. 

What is called 'photographic irradiation' affords one 
means of attack upon it. This arises from the diffusion of 
light within the substance of the gelatine film. The particles 
directly meeting the stellar rays reflect them irregularly all 


round to other particlee, thns widening the area of chemical 
decomposition, and creating circular images which, with the 
same exposure and on the same plate, are found to vary in 
size with the magnitudes of the stars they represent.^ Thus, 
from a few stars of ascertained brightness, that of the rest 
imprinted with them may, down to a pretty low grade, readily 
be inferred. The faintest stars, however, give rise to dots so 
small that their differences cannot be reliably measured. 

The method of 'trails* is a useful adjunct. For this 
purpose, the stars, instead of being held, so to speak, in a 
vice, by the following motion of the telescope, are allowed to 
travel across the plate, recording their passage in tracks of 
' reduced ' or metallic silver, the width and density of which 
serve as an index to the lustre of the objects originating them. 
But the almost instantaneous graphical power needed to pro- 
duce this effect belongs, as can easily be imagined, only to the 
brightest stars. Trails can practically be got from the fainter 
ones only in the immediate vicinity of the pole, where the 
diurnal motion is excessively slow.* 

For the lowest ranks of stars, the relative lengths of 
exposure needed to get impressions of them furnish the most 
promising means of determining their comparative light. 
The principles upon which this can be done are however still 
unsettled. The times, which for moderately bright stars 
sensibly follow the simple proportion of luminosity, do not 
appear to do so for fainter objects. A star of the thirteenth 
magnitude, for example, requires more than twice and a half 
times as long an exposure (other conditions being identical) to 
make itself perceptible as one of the twelfth. Exactly how much 
more must be decided by experience. A triple exposure for each 
additional magnitude has been provisionally adopted at Harvard 
College ; but no rule in the matter has yet been fully tested. 

Photographic is not always the same thing as visual 

' Astr, Nach. No. 2,884. See also, Charlier, Publication der Astr. Oesell- 
schafty xiz., Leipzig, 1889 ; Sohaeberle, Publications Astr. Society of the Pacific, 
No 4, p. 61 ; Holden, ibid. No. 5, p. 112. 

* By changing the rate of the clock so as to let the telescope lag slightly 
behind the stars, the effect of a diurnal motion as slow as desired can be 
given; but the trails will be proportionately shortened with the same exposures. 


brightness. Colour interposes a distinction ; since the quick 
vibrations at the blue end of the spectrum are those chiefly 
effective in releasing silver from chemical bonds. Blue stars 
are consequently far more, and red stars far less conspicuous 
self-printed than to the eye. Photographs of tinted double 
stars thus often show curious reversals, a small blue com- 
panion coming out superior to its yellow or reddish primary.' 
And stars too faint for chromatic discernment with the tele- 
scope can sometimes be picked out on a negative as coloured, 
simply through discrepancy of relative magnitude. These cases 
are, however, exceptional. In general, the eye and the sensitive 
plate agree in estimating gradations of stellar Ught. 

Nebular photometry is still in its infancy. Dr. Huggins 
ascertained, in 1866,^ the extreme intrinsic faintness of such 
objects; and there the matter rested until the universal 
agency of the camera was made available. At a meeting of 
the Boyal Astronomical Society on March 8, 1889, Captain 
Abney described a mode of obtaining a scale of photographic 
density by exposing to a standard light during different inter- 
vals of time, a number of small squares on one corner of a 
prepared plate. These, developed with the subsequently taken 
picture of a nebula, afford so many terms of comparison for 
the relative brightness of its parts, ^the effect of increase 
or decrease of time in determining density of deposit being 
exactly equivalent to increase or decrease of intensity of 
light.' ' A curve of varying lustre can in this way be drawn 
through the parts of a nebula-photograph, from a reference 
to which, at any lapse of years, it may be positively deter- 
mined whether local brightening or fading has occurred. 
Only internal luminous relations, however, are capable of being 
thus tested. The contrivance gives the means of comparing 
different sections of the same nebula, not one nebula with 
another, nor even the same nebula, as a whole, with itself at 
different epochs. Its adoption may nevertheless be expected 
to make a beginning in exactknowledge of the light-changes of 

I Instances are given by Mr. Espin, Observatory, vol. vii. p. 247. 
» Phil, Trans, vol. clvi. p. 392. 
• Observatory, vol. zii. p. 160. 


There is scarcely one of the namerous tasks of nebular 
astronomy that cannot be better performed photographically 
than visually. In the simple perception of faintly illuminated 
surfaces, the promptly fatigued living retina is left far behind 
by the imperturbable gaze of a sensitised plate; in their 
delineation, the subtlest human hand is at an equal, if not 
greater disadvantage. Spectroscopic inquiries, both stellar 
and nebular, are enormously facilitated by the substitution 
of permanent autographic records for sets of quivering hnes, 
measurable only at critical moments, and constantly liable to 
be effaced by atmospheric intervention. It is true that the 
range of observation is not the same in both cases. The 
plates now in ordinary use ignore the lower end of the 
spectrum, while reaching up much higher than the eye can 
follow into the ultra-violet. The use of coal-tar products, 
however, recently adopted in their manufacture, promotes 
sensitiveness to yellow and red rays ; and ' orthochromatic ' 
plates absolutely free from colour-preferences can be produced 
by special processes. So that even the part of the spectrum 
which seemed reserved for visual study will doubtless before 
long be appropriated by photography. 

The wonderful comprehensiveness and adaptability of this 
method are strikingly apparent in the results obtained of 
late years at Harvard College. By no other means could the 
spectroscopic stellar survey partially completed there, have 
been carried out on so great a scale. Since 1885, 28,000 
spectra of 11,000 stars have been catalogued; and similar 
data for southern stars are now being collected at Peru with 
' the same eight-inch ' doublet ' previously in use at Harvard. 
The mode of work (though suggested by Father Secchi) is a 
practical novelty. A prism large enough to cover the entire 
object-glass is placed in front of it. The stellar beams are 
thus analysed before they are concentrated; every stellar 
image is transformed into a prismatic riband ; and stars by 
the dozen or by the score print their spectra together on a 
single plate with a single exposure. Slit and cylindrical lens 
are alike rejected ; the diurnal motion is employed to widen 
the spectral bands suflSciently to bring out their distinctive 


features. That is to say, the stars are allowed to ' trail ' slightly 
across the direction in which their light is dispersed. The 
results are admirable; innumerable lines are clearly recorded ; 
but the difficulty of measuring them in the absence of any 
system of reference-lines has not yet been fully overcome. 

For detailed identifications, the more laborious plan 
adopted by Dr. Huggins in 1879 is still pursued. The stars 
are taken one by one ; their rays are admitted through the 
postern-gate of a slit, and record their peculiarities side by side 
with a comparison-spectrum providing starting-points for 
measurement. Dr. J. Scheiner is thus, at Potsdam, engaged 
in collecting data of hitherto unapproached accuracy from the 
same plates used by Dr. H. C. Yogel for determining stellar 
motions in line of sight. In this last difficult branch the 
superiority of the photographic method is perhaps more con- 
spicuous than in any other. The extraordinary precision at- 
tainable with it is chiefly due to the virtual elimination of the 
effects of air-troubles. The lines from which information as 
to movement has to be gathered depict themselves in their 
extra-atmospheric places. Their waverings, so baffling to the 
eye, are left comparatively unnoticed by the sensitive plate. 

Nor is it only here that the chemical mode of procedure at- 
tains a refinement on a par with its power. The subtlest pro- 
blem of stellar astronomy is that of annual parallax. Nowhere 
is there less room for compromise in the matter of accuracy ; 
yet it has been triumphantly solved with the aid of the camera. 
The issue of Professor Pritchard's careful and persevering ex- 
periments has been to render photographic determinations of 
parallax of equal authority with visual determinations, and so 
to quicken incalculably the rate of progress in collecting results- 
It happens moreover that the objects most inviting to the one 
mode of treatment are precisely those reached with difficulty 
by the other. Stars too faint for the eye to deal with satis- 
factorily, come out on the plate in neatly measurable form, much 
more promising for exactitude than the distended images of 
brighter stars. Photographic parallax work has not, up to 
this, been attempted elsewhere than at Oxford. 

It has been executed by means of a large reflector (thirteen 


inches in aperture), the gift of the late Dr. Warren de la Bae, 
the first example being thus given of the employment in tasks 
of such delicacy of that class of telescope. For picturing pur- 
poses, indeed, mirrors have the special advantage over lenses 
of being perfectly achromatic ; they collect in one focus aU the 
rays, visible and invisible, striking them. But even in the 
very best refractors, this is not the case. However skilful the 

* achromatic ' combination of different kinds of glass, a large 
amount of light is necessarily ' thrown away.' ' Opticians have 
to choose what sections of the spectrum they will turn to 
account, and neglect the rest. Photographic refractors are for 
this reason useless in ordinary observation. The images they 
give are wholly built up out of blue light, while the light 
proper for seeing by wanders unserviceably astray. Hence 
the plates exposed with them must be of an approximately 
uniform character. No tolerable results could be got with 

* orthochromatic ' plates in the Henry telescope. These evils, 
should the new Jena glass prove a workable reality, will be 
greatly diminished, but they will not be abolished. 

They are, however, to a great extent outweighed by 
countervailing advantages. Befractors are more manageable 
than reflectors. They are less sensitive to slight strains, less 
intolerant of unequal pressures ; they accommodate them- 
selves better to mechanical exigencies, can be more rigidly 
mounted, hence made to follow more strictly the circling of 
the sphere, and so to keep a steadier hold of the objects in the 
field of view. Where quantitative precision is chiefly aimed at, 
choice is thus naturally directed to them, and they have been 
stamped as the official instruments of celestial photography by 
their adoption for the vast star-charting operations decided 
upon at the Paris Congress. The splendid nebular delinea- 
tions, on the other hand, achieved with reflectors, by Messrs. 
Common and Boberts and by M. von Gothard, strongly re- 
commend their application to that and other special purposes. 
Each telescopic genus thus has its place in the boundless 
fields of photographic research; each has its own line of 
superiority ; neither excludes the other. 

» Sir H. Gkubb, Monthly Notices, vol. xlvii. p. 809. 


The fonndation of stellar astronomy is, as we have said, in 
infinitesimal accuracy. It could not otherwise exist, since the 
quantities concerned are so small as to get buried out of sight 
amid the errors of rough observations. But for its progress, 
something more is required. A few scattered items of know- 
ledge do not constitute a science. The word implies the suffu- 
sion of a subject with intellectual light derived from large in- 
ferences. Large inferences must, however, be based on a wide 
store of facts ; and as yet the facts collected by sidereal study 
are few indeed compared with its innumerable objects. Beason- 
ing is therefore cramped in cautious minds, or left to run wild 
in impatient ones, for lack of data. They are needed to wing 
thought in the one case, to restrain it in the other. And by 
photography alone it would seem that they can be supplied in 
the abundance and with the promptitude required. For, to a 
certain extent, the work has to be done against time. Just 
as rapid intuitions are necessary for following a long train of 
mathematical reasoning, because, where the steps are laboured, 
the wearied faculties at last refuse to continue to take them, so 
some degree of forward impetus is indispensable for sustaining 
the universal interest which gives a subject its vitality, but 
declines to follow too tardy an exchange of one halting-place 
for another. 

Thus, not merely what it can do, but the rate at which it 
can do it, has to be considered in estimating the value of 
photography as an ally to astronomy. And there is little 
fear of its admitting lassitude through sluggishness of pace. 
It keeps up a very ' Sturm und Drang ' of progress. The 
decuple powers of enumeration desired by Homer for cata- 
loguing the crowd of Greek ships are far outdone by it. Its 
instrumentality, moreover, came to hand just when the multi- 
tudinous character of the problem set by the heavens began 
to be grasped in all its formidable reality. 

The swiftness of the photographic method is due, not 
alone to the great number of objects it can register together, 
but to the dispersion and division of labour it makes possible. 
Becords obtained by it have the enormous advantage of being 
permanent. They fix the flitting incidents of the heavens as 


the phonograph fixes the transient accents of the human 
voice. All the scanty hours of unclouded darkness can accord- 
ingly be devoted to securing materials for subsequent inves- 
tigation in daylight or bad weather. Innumerable experts may 
be employed in this way at remote places and with different 
ends in view ; and a single negative might conceivably serve 
as the basis for successive inquiries in stellar photometry, 
chromatics, distribution, parallax, and proper motions. The 
danger seems rather to be that star-prints may get too little 
than too much study ; nor will zeal in securing them avail aught 
without industry in discussing them. The possession of pictures 
of celestial objects does not in itself constitute an increase of 
knowledge. They contain latent iniormation just as the skies 
themselves do, but the educing process by which it is made 
sensible is as necessary in the one case as in the other. 
Hence the visible rise into importance of a new branch of 
practical astronomy which might be designated ' astrometry,' 
and the obvious likelihood that significant detections will in 
future more and more be made with the microscope in the 
study, by investigators who have perhaps never in their lives 
worked with a telescope. 

Not that the telescope is, or ever can be, superseded. On 
the contrary, the enlargement of its capacities becomes more 
desirable with every fresh addition to the apparatus used in 
conjunction with it. Modern sidereal astronomy may be said 
to live on light. Large telescopic apertures are a sine qud non 
for its growth and activity. Indeed, the objects it has to do 
with are, in great measure, otherwise invisible. 




The stars, speaking broadly, are suns. But what is a sun ? 
We can only reply by taking Junction into consideration. A 
sun is a great radiating machine, and the obvious criterion 
for admission to the order is fitness for this office. Qualifi- 
cation to be a centre of light and heat is the dominant 
characteristic of each of its true members. Now the solar 
emissive activity is concentrated in a shining shell of clouds 
known as the * photosphere,' which the entire energies of the 
orgcmiam (so to speak) seem directed to maintain and renew. 
And with reason, since its efficiency as a radiator depends 
upon the perpetuation of the condensing process by which 
this brilliant surface is produced. 

The possession of a photosphere must then be regarded as 
an essential feature of the suns of space. But such a struc- 
ture can only be formed in an incandescent atmosphere, the 
action of which modifies, more or less powerfully, the light 
emitted from it. The spectroscope can then alone decide 
whether a given sidereal object be, in the proper sense, a sun. 
For it is not so much the quantity as the quality of its radia- 
tions that determines the point. They must be such as can 
be supposed to emanate from condensed and vividly glowing 
matter bathed in cooler, though still ignited, vapours. That 
is to say, they must be primarily unbroken from end to end of 
the rainbow-tinted riband formed by prismatic dispersion, 
while showing the secondary effects of absorptive encroach- 
ments. A continuous range of vivid light derived from a 
photosphere, crossed by dusky lines due to the atmosphere 
through which the photospheric radiations have to make 


their way out into space, is hence the distinctive spectrum of 
a snn. 

The enormous h'ght-power, and, so fax, the solar nature of 
the stars, followed as a corollary from the Copemican theory, 
since at the unimaginable distances implied by their apparent 
immobility while tiie earth performed its vast circuit, they 
should otherwise have been totally invisible. But the analogy 
could be strictly tested only by spectrum analysis, and it 
proved practically complete. Complete, that is, for the great 
majority of the stellar populace. There is a residuum in 
which it is impaired ; there are a few instances in which it 
is actually overthrown. 

This degradation of type shows itself in different ways. 
Absorption in some cases becomes so inmioderate as well-nigh 
to smother the original light of the star ; the atmosphere in 
others outshines the photosphere, giving rise to bright instead 
of dark lines in the spectrum, while a similar effect may at 
times be produced rather by a decline of photospheric, than 
by a heightening of atmospheric radiative intensity. When 
the failure has gone so far that the light of a seeming star, 
analysed with the spectroscope, is found to consist chielSy of 
isolated rays of various colours, then the object approximates 
more to a nebula than to a star. It certainly has no claim 
to the designation of a sun. 

But as in the other kingdoms of nature, so here ; there are 
no abrupt transitions. Continuity is everywhere maintained. 
The descent from a perfect sun to an undoubted nebula is 
effected without interruption. "We propose, however, in the 
present chapter, to consider only bodies of assured status, with 
radiative machinery in full working order — bodies, as to the 
essentially sun-like nature of which there can be no difference 
of opinion. Their continuous spectra are crossed, more or 
less numerously, by dark lines serving as the * recognition- 
marks ' of the various chemical substances ignited in their 
atmospheres. They are inscribed from end to end with hiero- 
glyphics which it is the chief business of the physical student 
of the stars to decipher. 

About eleven-twelfths of all the stars show linear spectra 


of absorption. They fall into two great divisions, correspond- 
ing to Father Secchi's first and second spectral types, which 
include respectively ' Sirian ' and * solar * stars. 

The Sirian stars are of a brilliantly white colour, sometimes 
inclining towards a steely blae. Sirins is the exemplar of the 
class, but at least every alternate star in the sky is of analogous 
constitution. Among the more conspicuous examples may 
be mentioned: Vega (a Lyrse), Algol 08 Persei), Canopus, 
a Crucis, /3 Argus, Spica {a Virginis), Regulus (a Leonis), 
Castor (a Geminorum), /3, 7, S, s, 17 UrssB Majoris, 

The light of these objects is not materially encroached 
upon by absorption. There is little or no trace in them of 
the general damping-down through vaporous intervention, by 
which our own sun is shorn of a large proportion of his more 
refrangible beams. The Sirian spectra, although not intact, 
are entire, and are hence especially strong in their ultra-violet 
sections. To the feebleness in them of absorptive effects 
there is, indeed, one remarkable exception. The sign-manual 
of hydrogen is stamped upon them with extraordinary inten- 
sity. By their broadened and hazy aspect the lines charac- 
terising this substance testify to its existence in these stellar 
atmospheres at an exalted temperature, and under pressure 
relatively high, yet certainly inferior to that of the terrestrial 
atmosphere at sea-level. Besides hydrogen, a number of 
metals are present, but, it would seem, in insignificant pro- 
portions, since they show lines too faint and fine for easy 
recognition. Fraunhofer's * D,' the ubiquitous double-line of 
sodium, is nevertheless unmistakable, as well as the magnesium 
group &, and a line of iron in the green part of the spectrum 
known as *E.' 

The most striking feature in spectra of this type has, how- 
ever, been disclosed by photographic means, for the application 
of which the strength of their ultra-violet radiations peculiarly 
fits them. These are powerfully absorbed by glass, and Dr. 
Huggins accordingly rejected this material in toto from the 
apparatus employed by him in the autumn of 1879.* All the 
lenses included in it were made of quartz ; a metallic mirror, 

' * > Phil, Trans, vol. olzzi. p. 669. 



eighteen inches in diameter collected the light afterwards 
sifted by transmission through an Iceland-spar prism. A 
discovery of singular interest resulted. 

A photograph of the spectrum of Vega obtained with one 
hour's exposure contained twelve strong lines (shown in the 
accompanying illustration) forming a group in obvious rhyth- 
mical connection, and crowding together in an accelerated 
proportion as their wave-lengths shorten. A common origin 
for the entire series at once suggests itself : and, on the ground 
that its two most refrangible members were already known as 
hydrogen-lines, Dr. Huggins did not hesitate to pronounce 
hydrogen responsible for all. His inference has since been 
amply justified. Seven of them proved to have been, a short 

Fig. 1.— Spectrum of Hydrogen in White Stars compared with the Spectrum of 
the Orion Nebula (Huggins). 

time previously, photographed du-ectly from glowing hydrogen, 
by Professor H. W. Vogel ; * and the entire series was similarly 
procured later by M. Cornu K But with extreme diflBcuIty. 
Although the purified gas filling the capillary tube placed in 
front of the slit was excited by powerful electrical discharges, 
its highest radiations took no less than three hours and a half 
to get satisfactorily printed. Some idea may thus be gained 
of the intense incandescence reigning in the remote stellar 
atmospheres from which they were first derived. 

The complete spectrum of hydrogen comprises fourteen 
lines, the designations of which, with the actual lengths of the 

» Astr. NacK No. 2301. * Journal de Physique^ t. v. p» 841 (1886). 


ethereal waves producing them, measured in ten-millionths 
of a millimetre, are subjoined as they appear in fig. 1, irom 
which, however, the least refrangible, named C,iB omitted. as 
lying too far to the right. 

In following the course of sidereal research, it is important 
to bear in mind at least the order of their succession in the 
spectrum. The two first, belonging to its visible part, are, it 
should be remarked, evidently correlated by their wave-lengths 
with the others, so as to form part of the same progression. 



1 Frannhofer'E 

1 C 





8 near „ 









6 Hnggins's 


























The fact is a memorable one that the true character and 
full extent of the hydrogen-spectrum became known through 

* astro-physical ' inquiries. It shows with what curious 
unexpectedness the obligations of one science to another may 
be repaid, and exemplifies the advantages to be reaped by 
terrestrial chemistry from extending its experimental range to 
the heavenly bodies. The significance, for its purposes, of the 

* white star ' set of rays has been heightened by Cornu's 
detection of a corresponding series in the ultra-violet spectra 
of thallium and aluminium.* Not only is the title of hydrogen 
to rank as a genuine metal thereby ratified, but the existence 
is hinted at of a general law of spectral emission which will 

* Journal de Physique^ t. v. ser. ii. p. 93. 


perhaps eyentually prove commimicatiye regarding the coveted 
secrets of molecular structure. 

The complete presence of the hydrogen series may be 
called the badge of the first type of stellar spectra. ' No 
celestial body/ Mr. Lockyer remarks, ' without all the ultra- 
violet lines of hydrogen discovered by Dr. Huggins, can claim 
to belong to it.' ^ Accidental circumstances, it is true, may 
encroach upon the integrity of the set in any given photograph. 
Either through atmospheric hindrances, or through inadequacy 
in the supply of light, partial failures must be expected in 
attempts to get impressions of such susceptible vibrations. 
Thus, in the spectrum of Sirius, which comes next to that of 
a LyrflB in fig. 2, the absence of the five uppermost lines is 
entirely due to the slight elevation above the horizon of the 
star when photographed. Becalling, indeed, that this series 
has only by straining the resources of the laboratory been 
induced to print itself from a Ught-source close at hand, we 
can but marvel that its absorptive effects are to any extent 
perceptible in stellar beams attenuated by distance, and com- 
pelled, at the end of their journey, to submit to atmospheric 
incursions especially directed against their ultra-violet con- 

All the six ' white-star ' spectra then, represented in fig. 2, 
may be considered to bear, in reality, the full stamp of the 
emanations of hydrogen. Casual deficiencies are invari- 
ably of the most refrangible lines ; they occur in the region 
where want of Ught first begins to tell, and just when the 
spectrum becomes characterless through faintness. 

There is hence reason to believe the presence of the com- 
plete absorption-spectrum of hydrogen to be a searching test 
of stellar constitution. The stars in which it is apparent 
are so closely allied as most properly to form a class by them- 
selves. It includes, however, a variety, of which the chief 
stars in Orion (Betelgeux excepted) are the representatives, 
distinguished by Yogel for the almost total blankness of their 
visible spectra. Tet the photographic investigation of their 
light has not proved barren. All the typical lines of hydrogen 
* Proceedings Royal Society, vol. zly. p. 882. 

5 « OL) « -«■ < 

M 1^ I I I 

5 i ei P ? 3 

« p" « « d 

S • 

- fed! 

a li n It 

Fio. 2.— Photographed Stellar Spectra (Hoggins) 


appeared in Dr. Huggins's photograph of the spectrum of 
Bigel ; and Dr. Schemer, from an examination of the section 
F to H, to which the impressions obtained at Potsdam are 
limited, derived some interesting conclusions respecting the 
stellar species exemplified by RigeL* (See fig. 3.) All the 
lines in the spectra of these stars — and there are many be- 
sides those of hydrogen— are equally broad ; and they have 
the peculiarity of being broad without being hazy. The ex- 
planation suggested is that the absorptive action producing 
them takes place in a cooler and thinner atmosphere than 
that of ordinary first-type stars ; but this is doubtful. Dr. 
Scheiner found the spectrum of a Cygni, while visually of the 
pure Sirian type, to be photographically akin to that of Rigel ; 
and he identified in it a multitude of iron-lines showing 
curiously different relative intensities from those belonging 
to the same lines in the solar spectrum, and originating, 
it was accordingly thought, under abnormal conditions of 
temperature.^ In the spectrum of Sirius, forty-three out of 
ninety-one fine lines measured in the space P to H are cer- 
tainly due to iron. The drawings reproduced by Dr. Scheiner 's 
kind permission in fig. 3 exhibit, with close fidelity, the photo- 
graphed spectra of Capella, Altair, Sirius, and Rigel, over a 
third of their length, but on an enlarged scale. 

Although the hydrogen spectrum is dominant throughout 
the first order of stars, it is not in all represented with equal 
emphasis. The diffuse lines constituting it in Sirius and 
Vega show, in descending gradations of fineness, in Benet- 
nasch (17 Urs8B Majoris), Spica, Altair, and a Cygni (see fig. 
2). They appear, moreover, less solitary in proportion as 
they become less intense. Their possession of the ultra-violet 
field, all but exclusive in Sirius and Vega, is progressively en- 
croached upon, in the succeeding stars, by the development 
of other spectral lines. The co-ordination of the two kinds 
of change may be expressed by the general statement that 
the consjpicuousness of rays due to absorption by ordinary metals 
in the spectra of white stars varies inversely with that of the 

> Asir. Nach. Nos. 2923-4. 

' Siteungsberichte Akad. der Wissenschaftefif Berlin, 1890, vol. viii. p. 7. 


a Aorigae. u Aqiiilie. a Cun. MhJ. $ Orionis. 

Fio. 8.~ Photographed Stellar Spectra (Soheiner). 


hydrogen aeries. When the hydrogen-rays become e£EiEkced 
from the invisible, and cease to be predominant in the visible 
part of the spectrum, the second, or solar type of stars is 

These are about one-sixth less numerous than the first 
kind. We may take as examples : Capella, a UrssB Majoris, 
a Cassiopeise, a Arietis, s Argus, a Serpentis, Aldebaran, and 
Arcturus. The pole-star, Procyon, a AquilsB, and probably 
the brilliant southern binary a Centauri, stand nearly mid- 
way between the two groups. 

A golden tinge like that of sunlight betokens, in stars of 
the second order, a spectrum more or less perfectly similar to 
that of the sun, delicately ruled from end to end through the 
absorptive eflfects of a great variety of metallic vapours ; non- 
metallic substances give no sign of being present, unless by 
a trace of carbon in the sun. The extent to which the lines 
are crowded together may be judged of from Cornu's photo- 
graphic solar spectrum depicted in the upper horizon of 
fig. 2. Dark hydrogen-rays are present, but with no pre-emi- 
nence, and the series is apparently not continued above h in 
the violet. We say * apparently,' because the fifth hydrogen 
line may lurk concealed within the shadow of an obscure 
diffuse band which covers its place. This band was named 
by Fraunhofer H, and its companion, a little higher up, is 
designated K. The pair form the most strongly-marked 
feature of the spectrum of calcium when raised to the highest 
pitch of incandescence ; and there is much to be said in favour 
of Mr. Lockyer's view that they emanate, not from calcium 
in its entirety, but from some of its subtler ingredients. There 
is at any rate no doubt that they are what is called ' high 
temperature lines ' ; the light of ordinarily glowing calcium 
does not contain them. 

In their * unreversed,' or bright condition, they are 
peculiarly characteristic of the spectrum of solar prominences. 
Daylight observations at the edge of the sun show them 
always brilUant in the chromosphere, and often extending 
to the very summits of its flame-like extensions.* During 

* Toung, Natttre, vol. zziii. p. 281. 


the total eclipse of May 17, 1882, the violet radiance of H 
and E flooded the shadowed part of our atmosphere, and, 
dimly illuminating the purple disc of the moon, was scattered 
far out among the * aigrettes ' of the corona. 

Now, as we have said, the calcium H falls very near 
indeed to the hydrogen H. The difference of their wave- 
lengths amounts to no more than one ten-millionth of a 
millimetre, and it is only under favourable circumstances 
that they can be seen as distinct. As a rule, either the 
hydrogen-line is so widened as to mask the calcium line, or 
the calcium-line as to enwrap the hydrogen-line. Professor 
Young, however, saw them side by side, bright, during his 
observations of prominences in 1879-80 ; and they appear 
dark in a photograph of the spectrum of a Cygni taken at 
Harvard College, November 26, 1886. In this star, the 
hydrogen rays have thinned down almost to their solar con- 
dition, while other metallic lines are fine, yet pronounced. 
Such a critical balance of conditions alone made possible the 
separate discernment of the two lines on the Harvard plate. 

Thus, while the double origin of H opens the way to mis- 
understandings of its true significance, the state of the 
calcium-line at E is a most useful index to the physical con- 
dition of a star. Next to the mode of appearance of the hy- 
drogen series, it is perhaps the most significant individual 
feature in analysed star-light. The substance emitting both 
H and E is evidently of first-rate importance among the 
vapours surrounding the sun. But, so far as terrestrial 
experiments can inform us, it arises as a modification of the 
metal calcium only under the strongest electrical excitement. 
This no doubt generates an enormous degree of heat, but is 
perhaps not identical with it in its method of action upon 
matter. Professors Liveing and Dewar advert to the probability 
' that the energy of the electric discharge, as well as that due to 
chemical change, may directly impart to the matter affected 
vibrations which are more intense than the temperature alone 
would produce.* ^ There is a practical certainty that, under 
some circumstances, bodies are rendered electrically luminous 
■ Proceedings Boyal Society, vol. zliy. p. 24a« 


at comparatively low temperatures ; it would be miphilosophical 
to assume that, under others, of which we know little or 
nothing, the distinction ceases to be valid. The fact, however 
we view it, is undoubted, that E is almost effaced from the 
spectra of what we have reason to believe are the hottest stars, 
while devdoping remarkably in cooler bodies. It may be that 
Dr. Huggins's proposed arrangement of photographed stellar 
spectra in a continuous series, according to the character in 
them, or even total absence from them of this line,' has a 
deeper meaning than is at once apparent. 

Considerable diversity of detail is to be found among 
' spectra of the second type. The model solar star is Capella ; 
in the invisible, as well as the visible part of its prismatic 
light, all the characteristic solar groups exist in about their 
\ solar strength. Dr. Scheiner identified with extreme precision 
255 lines photographed from Capella (between wave-lengths 
4124 and 4688), with lines derived from the sun* (see fig. 8) ; 
and there is every indication of an almost perfect constitutional 
similarity between the two bodies. 

In Aldebaran this standard is pretty widely departed from. 
The pale rose tint of its light is accounted for by the slightness 
of absorptive effects in the red end of its spectrum, while 
numerous lines modify the yellow and green, and the violet 
rays are so feeble that, with an exposure ffty times that re- 
quired for Sirius, Dr. Huggins obtained only an impression 
virtually bounded by P and H. This feebleness of chemical 
action may, however, be due, not to original deficiency of 
light, but to its stoppage in the vaporous envelope of the 

Of its spectral lines, seventy were determined by Huggins 
and Miller in 1868,' seventy-two by Vogel ten years later,* 
fifty-four being clearly traced by both observers to the radia- 
tions of the following nine substances : hydrogen, sodium, 
magnesium, calcium, iron, bismuth, tellurium, antimony, 
and mercury. Beyond a doubtful indication of bismuth, there 

» Phil. Tram. vol. dxxi. p. 679. » Ast/r. Nach. No. 2928. 

■ PhU. Trans, vol. cliv. p. 424, 

* Bothkamp Beobachiungen, Heft ii. p. 11. 


is no sign in the solar spectrom of the arrest of light by any 
of the last four elements,^ the high vapour-densities of which 
it is important to note. 

K we were to consider only the visible spectrum of Arcturus, 
we should certainly suppose it to vary from the sun in the 
same direction as Aldebaran — that is, to be farther removed 
than Capella from the Sirian stars. Its rays, like those of 
Aldebaran, have a ruddy tinge, and are seen, when dispersed 
by the prism, to be powerfully stamped by metallic absorption. 
But the camera has a different story to tell. In the intensity 
of its ultra-violet as compared with the rest of its beams, 
Arcturus falls but little behind Sirius. Their inscription with 
legible characters extends, indeed, very much higher up (see 
fig. 2).* The characters, however, are unlike what might have 
been expected. Up to the limits of visibility, the solar analogy 
is fairly well preserved, H and K (the latter especially) being 
even more distended than in the suii. But beyond, non-solar 
groups of strong lines appear ; and among them, six out of 
the nine typical rays of hydrogen. 

The star with which Arcturus shows the closest analogy 
in the ultra-visible part of its spectrum, is a AquilsB. And it 
is remarkable that the resemblance concentrates itself in 
features common to solar appendages and to the two stars. 
The spectrum of the ' Sohag prominence,* photographed by 
Dr. Schuster in 1882, contained, besides H and E and the 
complete hydrogen-set, a number of bright lines of unknown or 
doubtful origin. In the course of his examination of the plate, 
Captain Abney was struck with approximate coincidences 
between ten of them and dark lines photographically recorded 
by Dr. Huggins from a AquilsB.* Seven of these common 
lines reappear in Arcturus, besides five others, probably albo 
identifiable with prominence-rays. The preponderance of the 
violet calcium pair (H and E) is a further point of agreement. 

• Lockyer, Chemistry of the Sun, p. 220. 

' Dr. Hoggins obtained, however, in the spring of 1890, a photograph of 
the spectrum of Sirias showing a group of six broad lines more refrangible 
than any of the hydrogen series. Their approximate wave-lengths are 3838, 
3311, 3278, 8254, 3226, 8199 ten-millionths of a millimetre. 

• Phil. Tram. vol. chcxv. p, 267. 


The peculiarity of the analysed light of Arcturus consists, 
then, largely in its inclusion of the prominence-spectrum 
strongly 'reversed/ or turned from bright to dark. This 
seems to tell two things : first, that this star has very extensive 
chromospheric surroundings — ^that it possesses an enhanced 
equivalent of the ocean of incandescent gases making a rose- 
red edge to the black moon during total solar eclipses. Next, 
that it is a hotter star than the sun. We must, however, 
define what we mean by saying that one star is ' hotter ' than 
another. The expression, otherwise open to miBunderstanding, 
conveys, as used here, simply that the temperature of the 
emitting surface is higher in one case than in the other. The 
photosphere is at an intenser glow ; the heat is more concen- 
trated at a particular level. 

But, it may be asked, how can we judge of the temperatures 
of the stars ? By what criterion can we compare them ? Not 
either by colour, or by the relative extent of their spectra, 
since both depend upon the kind and amount of absorption 
produced by their atmospheres, more than upon the original 
quality of their light. To describe, for instance, the white 
and red stars as illustrating respectively the white and red 
stages of heat, would mislead by the use of a totally irrele- 
vant simile. Criteria of other kinds enable us, nevertheless, 
to discriminate, as to temperature, between star and star. 
The surest perhaps are those relating to ' reversals.' 

The display of dark lines upon a bright background 
implies that the continuous light emanates from a source 
hotter than the interposed absorbing vapours. Thus, the 
photospheres of Sirian stars must (so far as we can see) be at 
higher temperatures than the hydrogen-strata emitting the 
series of vibrations thrown, as it were, in shadow upon them. 
For, if the strata were hotter, the lines would show bright 
against the photospheric radiance; if they were of equal 
temperatures, the lines would not show at all, radiation just 
balancing absorption. Now the ultra-violet members of the 
series, although they can be derived directly from the solar 
surroundings during total eclipses, make no mark in the 
solar spectrum. And why ? Either because the solar photo- 


sphere is not hot enoagh to reverse them, or because the solax 
chromosphere, in the main, is not hot enoagh to emit them. 
Both conditions are, however, combined in the Sirian stars 
as well as in Arctorus, the glow of all of which is hence pre- 
sumably more fervid than that of the son, Capella, or Alde^ 

The spectra of several of the Orion stars are beheved to 
be variable.^ Visual examination, at least, sometimes wholly 
fails to show the hydrogen-lines in Bigel, while at other times 
they are distinct, even conspicuous ; nor are indications alto- 
gether wanting of relative fluctuations affecting C and F.' 
Analogous changes have been recorded by M. von Konkoly in 
^ Ononis, the third star in the belt ; while its companions 
s and S wear, occasionally, by his observations, the aspect of 
solar stars. 

The two first stellar orders are thus probably connected, 
not only by gradation but by migration. The dividing-line, 
rendered difficult to draw by the occurrence of intermediate 
examples, is still further effaced by the swinging across it of a 
few unstable objects. Light, however, is not darkness, because 
each melts into twilight. Distinctions are none the less real 
for being estabUshed by almost insensible transitions. Let us 
recall what the distinctions are. 

Spectra of the Sirian pattern serve, it might be said, as 
brilUant screens for the display of the reversed hydrogen rays. 
But only four out of the fourteen appear, and those com- 
paratively shrunken and insignificant, in Capella and the sun. 
Mixed absorption, which in Sirian stars is insignificant, has 
instead become predominant. The hght of Vega, for instance, 
reaches the outskirts of our air substantially as it left the 
star's photosphere. It retains its native bluish tinge; the 
proportionate intensities of its variously refrangible portions 
remain unaltered. But the sun, if its vaporous envelope 
could be suddenly exchanged for that of Vega, would probably 
leap up to three or four times its present lustre. Its rays, 
no longer subdued into benignity, would have a keen edge 

» Vogel, Astr, Nach, No. 2889. 

' Gyalla Beobachtungmt Bde. viii. p. 5, x. p. 57. 


to them, and would dazzle like lightning with their violet 

What kind of change, we may now ask, in the physical 
condition of the sun might conceivably produce this effect ? 
Under what circumstances can we suppose its investing strata 
reduced to consist almost exclusively of hydrogen ? Certainly 
not by a simple increase of heat, which, through the resulting 
more copious metallic vaporisation, should prodiit^e, other 
things being equal, a more complex, and presumably a more 
strongly absorptive atmosphere. An increase of gravity, were 
that possible, would come nearer the mark. For it would 
deprive heavy metallic vapours of power to rise to any con- 
siderable distance above the photosphere; most of them, 
indeed, would probably sink altogether below it ; the lighter 
ones that remained above, but close- to it, would, on that 
account, be too hot for the effectual stoppage of its light. 
Hydrogen alone, owing to its enormous elasticity, would still 
float in the cooler regions, though in layers very much more 
concentrated than at present. Hence the widening of its cha- 
racteristic lines. On such grounds as these the view was 
urged some years ago by Mr. Johnstone Stoney,* that white 
stars are actually differentiated from yellow ones by the 
stronger mastery in them of gravity over temperature. Just the 
reverse, however, of this appears to be true ; the mass of Sirian 
stars is probably small compared with their radiative energy. - 

But heat and gravity are not the only forces active in the 
neighbourhood of the sun. From a careful survey of all the 
evidence, Dr. Huggins infers the corona to be a product of 
electrical repulsion similar in nature to the tails of comets.^ 
And Professor Young points out that * no gaseous envelope in 
any way analogous to the earth's atmosphere could possibly 
exist there in gravitational equilibrium under the solar condi- 
tions of pressure and temperature.' ^ By the effects of their own 
weight, if unopposed, the vapours surrounding the sun, what- 
ever their temperature, should increase in density downward 

* Proceeditigs Royal Society ^ vol, xvii. p. 48. * Ibid. vol. xxxix. p. 127. 
' General Astronomy , art. 831. See also Prof. Bigelow's Solar Corona dis- 
cussed by Spherical Harmonics, Washington, 1889. 


with great rapidity; but rays as fine as if derived from 
a vacuum-tube are emitted by beds of them hundreds, nay, 
thousands of miles in thickness. It is certain, then, that 
gravity meets a counteracting influence, which is inexplicable 
except as the result of electrical repulsion. 

The various gradations of intensity of this counteracting 
force in the stars may serve in a great degree to explain the 
generic differences of their spectra. Thus its marked diminu- 
tion in the sun, equivalent to a virtual increase of gravity, 
would not improbably be attended by the transformation 
of the solar into a Sirian spectrum. The heavier metallic 
vapours would then altogether drop out of sight beneath the 
photosphere^ those that remained would exercise but a feeble 
absorption ; hydrogen would remain concentrated and pre- 
dominant. The corona, shorn of its rays and streamers, 
would appear, during eclipses, as a mere rim of vivid Ught ; 
the rosy prominences at its base would most likely have finally 
collapsed into a shallow chromosphere. The Sirian stars are, 
in this view, suns with very scanty appendages. They collect 
the whole of their luminous energies into the blaze of their 
photospheres. And it may be surmised that not so much the 
thermal as the electrical condition of stars is gauged by the 
fineness or strength of the significant calcium lines in the 
violet, which develop perhaps solely under the influence of 
powerful electrical excitement. 

E 2 




The shining face of each one of the innumerable suns aggre- 
gated into the vast system of the galaxy is, as we have said, 
veiled by absorbing vapours ; but in widely varying degrees. 
The light of Vega, though indelibly stamped with the charac- 
teristic lines of hydrogen, reaches our upper air without 
sensible general modification. Sunlight is not only charged 
with significant inscriptions, but throughout toned down and 
mellowed ; while in Aldebaran, the process of stoppage in the 
blue has been carried so far as to leave the red rays visibly 
predominant. In Aldebaran, too, the first symptoms begin 
to appear of a generic change in the manner of absorption 
through the emergence of dark bands in addition to the dark 
lines of the spectrum. 

This change is full of meaning. Isolated rays of definite 
wave-lengths, forming in the spectrum what we call * lines,' 
bright or dark, are emitted only at very high temperatures. 
They represent perhaps the fundamental vibrations of the 
' atoms ' of each different substance. But these, at lower 
grades of heat, are not free to thrill separately. Bound to- 
gether into * molecules,' they give rise, by their associated 
vibrations, to complex systems of Ught- waves, dispersed into 
sets of prismatic ' flutings,' each with a sharp and a nebulous 
side. The fluted spectra thus constituted are derived from 
chemical compounds such as oxides and chlorides, as well as 
from * elementary ' substances, both metallic and non-metallic, 
at moderate degrees of heat. But the flutings can always be 
made to give place to lines by sufficiently raising the tem- 
perature of the glowing vapour emitting them. 

It appears then that the increase of absorption in passing 
from the white stars through the yellow to the red, is attended 


a a certain point by the addition of another kind of light- 
stoppage, and one indicating the action of cooler vapours than 
those previously manifest. That is to say, a spectrum of 
bands or flutings is superposed upon a spectrum of dark lines 
similar to that of the sun or Aldebaran. The alteration 
obviously implies an augmented extent of absorbing atmo- 
sphere. For the bands must originate in a region of less heat, 
that is, at a greater distance from the stellar photospheres, than 
the lines. The stars displaying them are hence distinguished 
by the wide compass and varied heat^levels of their, incandes- 
cent surroundings, affording the means of arresting radiations 
by different systems of absorptive effects. And since the blue 
end of the spectrum invariably suffers most from transmission 
through vaporous strata, the red colour of these objects is easily 
seen to be a necessary accompaniment of their spectral 

Stars with banded spectra are of two distinct kinds, con- 
stituting Father Secchi's third and fourth stellar types. In 
Type III. numerous shadings terminate abruptly towards the 
violet, but very gradually towards the red, producing some- 
thing of a colonnaded effect. They are finely developed in 
Betelgeux (a Orionis), Antares (a Scorpii), 7 Cruois, aHerculis, 
/SPegasi, aCeti, Mira Ceti (variable), Lj Puppis (variable), 
&c. One star in four hundred may be roughly estimated to 
belong to this type. About 800 of them are at present known, 
of which not far from 800 have been discovered by Mr. Espin 
of Wolsingham with his ITJ-inch silver-on-glass reflector. 
Half a hundred have already been identified by photographic 
means in the southern hemisphere.^ 

These stars are invariably of a reddish or orange colour, 
very little of their blue light making its way through the 
piled-up vapours that surround them. Their spectra can 
hence be photographed only with great difficulty. With a 
forty-fold exposure (as compared with that needed for Sirius) , 
Dr. Huggins succeeded in getting from Betelgeux, the brightest 
star of the entire class, a bare trace of ultra-violet action.^ 

' E. C. Pickering, Henry Draper Memorial Fourth Anmial Report, p. 6, IBCO. 
^ Phil. Trails, vol. Ixxi. p. 886. 



Vogel's suspicion of the presence of hydrogen was, however 
confirmed by the appearance on the negative of a thin * G ' line. 
As a rule, no evidence of absorption by hydrogen can be found 
in star-spectra of the third type ; emission by it, however, 
leaves its mark in a few. This most universally diffused of 
substances is not then absent from such stars. It may exist in 
them as copiously as in the sun, yet by an approximate balance 
of temperatures neutralise in general its own action upon the 
light proceeding from their photospheres. Radiation and absorp- 
tion being thus about on a par, the marks of hydrogen vanish. 
In the visible spectrum of Betelgeux no less than ninety- 
five dark lines were measured by Huggins and Vogel,^ and 
these are only the more prominent among a crowd of others. 






B6 B5 B4 B3 B8 Bl 




■ 1 UUUi 



1 Hiiin 

H i V r- "■.' T ':? i |:r '1^ t5 '^ ? *=* t^ £ ^ i £: 5^ w ^ & ca 


Fig. 4.— Stellar Spectra of the Third Glass. 1, a Orionis ; 2, a Hercolis, 


The chemical substances indicated with certainty as consti- 
tuents of the star's atmosphere (besides hydrogen) are sodium, 
magnesium, calcium, iron, and bismuth ; silver, manganese, 
thallium, tin, cadmium, antimony, and mercury were more 
dubiously recognised. The vapour-densities of several of these 
metals are significantly high. In a Herculis, hydrogen makes 
no visible sign, but the spectrum is thronged with metallic 
rays of absorption, some cf them so strongly marked as to 
show, under Vogel's scrutiny in 1883 with the 27-inch Vienna 
refractor, across fluted shadings considerably denser in this 
star than in Betelgeux.* 

' PhU, Trans, vol. cliv. p. 426 ; Beobachtungen eu Bothkamp^ Heft i. p. 20, 
i. p. 16. 

* Bcoh. zu Bothkamp, Heft i. p. 28 ; Potsdam Publicationen, No. 14, p. 21. 


The two spectra can be compared in fig. 4, which repro* 
duces drawin«;8 made by M. Duner with the 9i-inch refractor 
of the Lund Observatory. The red end is on the right, as 
may be seen by the increase in that direction of the wave* 
lengths in the scale marked underneath. Of the few lines given 
in the figure, the majority, the reader will not fail to remark, 
appear to terminate bands. They seem placed there as if to 
accentuate the transition from brilliant to dusky tracts of 
prismatic light. Such coincidences are characteristic of the 
third type of stellar spectra ; they are too frequent, as M. Duner 
points out,* to be supposed casual ; yet they are totally unex- 
plained. The origin of the lines can indeed generally be traced 
either to iron^ ealcium, or magnesium : what is perplexing is 
that the edges of the bands should fall just in the same positions. 

The chemical meaning of the latter has not yet been satis- 
factorUy explained. The first point to be noted about them is 
their essential invariabiUty. All the leading bands are re- 
peated, as if in stereotype, in every star of the class. A few 
secondary shadings, it is true, are visible in some which can- 
not (owing perhaps to their inferior brightness) be distin- 
guished in others ; but the fundamental pattern of the spectra 
is everywhere the same.* So striking is this rigidity of design 
as to suggest that the seven or eight principal flutings repre- 
sent a single system of vibrations ; that they are due to the 
absorptive action of one substance. As a combined effect, the 
strict uniformity observed is extremely difficult to account for. 
There are, however, great differences, not only in the general 
intensity of the bands, but in their relative importance. Those 
numbered 2 and 8 in the figure are usually the most con^ 
spicuous, but numbers 7 and 8 predominate in some very red 
stars, the more refrangible rays of which are besides in large 
measure cut off by general absorption.^ 

The only systematic interpretation so far attempted of this 
remarkable class of spectra was offered by Mr. Lockyer, as a 
part of his * meteoric theory,' in 1887-8. Of the dark 
flutings they include, two are ascribed by him to absorption 

■ Sur Us ^toiUs a Spectres de la TroisiSme Classes p. 121. 
* Ibid. p. 8. » Ibid. p. 9. 


by manganese, two to magnesium, three more respectively to 
iron, barium, and lead.^ But these identifications depend 
upon coincidences of wave-length which profess to be no more 
than approximate, and they are hampered by apparently 
insuperable difficulties. They are coupled with another 
alleged feature in the spectra of these stars of still greater 
physical importance. The reader is aware that the distinc- 
tive light of comets consists mainly in the bright emanations of 
carbon, concentrated in a yellow, a green, and a blue band. 
These Mr. Lockyer asserts to be represented directly in the 
radiations of Betelgeux and its congeners. But the evidence 
that their spectra include the spaces of enhanced brightness 
which, if this were so, should just match the cometary flutings, 
is very far from conclusive. It consists almost wholly in the 
close agreement in position between the dark edge of a 
band in the stars (No. 7) graduated towards the red, and the 
briUiant edge of the green fluting in comets dying away 
towards the blue. The former, on Mr. Lockyer's hypothesis, 
is not a genuine band of absorption, but an effect of contrast 
produced by a bright carbon-fluting /ariw^ the other way. But 
the supposition is thereby involved that stars like Mira Ceti, 
in which this interval is almost black, are totally deficient in 
continuous light — a supposition contradicted by the photo- 
graphic, as well as by the visual brilliancy of their spectra. 
The invisibility of the yellow and blue carbon-flu tings,' always 
seen in comets, makes the presence of the green fluting appear 
still more improbable. 

The point is of crucial importance as regards the consti- 
tution of these bodies. In Mr. Lockyer's view, they are * not 
masses of vapour like our sun, but swarms of meteorites.' They 
are composed of millions of separate rocky or metallic fragments 
each circulating independently round the centre of gravity of the 
entire. Circulating independently, but with numerous jostlings 
and collisions, in the course of which heat is developed, and 

* Proceedings Royal Society, vol. xliv. p. 54. 

' There are grave objections to admitting the reality of the * masking* 
effects of manganese and lead absorption, by which Mr. Lockyer seeks to 
explain the non-appearance of the Anting at wave-length 5634. 


glowing vapours evolved. Each meteorite is encompassed 
with its own distinct little atmo- 
sphere, by the action of which cer- 
tain characteristic sections of its 
light are absorbed, while the * inter- 
spaces/ throughout the whole vast 
extent of the system, are filled with 
glowing hydrocarbons. Hence the 
supposed compound nature of a 
spectrum integrating the radiation 
of carbon and the absorption of 
manganese and lead,* and suggest- 
ing that the objects it characterises 
should be looked upon as comets 
on a prodigious scale, rather than 
as suns. 

But let us consider how the 
matter really stands. Beneath the 
series of shadings which diflFeren- 
tiate stars of the third from stars 
of the second class, lies a spectrum 
of lines of precisely the same gene- 
ral character with that of the sun, 
and bearing witness to a sun-like 
incandescence. Photographic testi- 
mony is strongly to the same effect. 
Fig. 5 shows the blue and violet 
parts of the spectrum of Betelgeux 
as imprinted upon a highly sensitive 
plate exposed at Harvard College. 
Like the solar spectrum, it is closely 
and delicately ruled throughout with 
fine lines of metalUc absorption, and 
bears no trace of radiation by carbon 
or any other vapour. The under- 
lying continuous light is obviously, 
as in the sun, photospheric — it 

' Proc. R. Soc. vol. xliii. p. IHO. 










is derived from the brilliant condensation-Burface of the star. 
Indeed, spectra of the second and third types are, in their 
photographic portions, so nearly identical, that difficulty has 
been experienced, in the course of the great cataloguing 
work carried out at Harvard College, in distinguishing them. 
Dr. Scheiner, too, remarks that all the chief lines visible between 
F and H in solar stars reappear in fluted spectra, with the 
addition, however, of numerous very narrow dark streaks 
diflfuse on one side, the strength of which seems to increase 
with the more pronounced development of the type.* 

In the manner of visibility of hydrogen in their spectra, 
however, these stars exhibit a marked deviation from the solar 
modeL Most, as we have said, give no sign of containing that 
element at all ; in one (a Ononis), it shows feebly obscure, 
while a few, all of them subject to great fluctuations of light, 
bear the stamp of its direct incandescence. Photographs of 
* Mira ' Ceti and of a similarly variable star in Orion (near ^ * 
Orionis), taken at Harvard College in the autumn of 1886, 
gave the earliest unquestionable proof of the occurrence of 
bright lines of any kind in banded stellar spectra of a normal 
type. We are enabled, by the kindness of Professor Pickering, 
to reproduce in fig. 6 a beautiful enlargement of the photo- 
graphed spectrum of Mira. It includes six vivid hydrogen- 
rays, namely, G, h, and four of the ultra-violet set. But the 
intermediate line at H is missing, its place being occupied by 
the closely adjacent calcium-line, which, with its comrade K, 
is just as dark, as broad, and as hazy as in the solar spectrum. 
There can be no question, however, but that the hydrogen- 
series is originaUy complete ; the H-Hne is radiated with the 
others, and would show equally bright were it not stopped-out 
by the overlying metalUc vapour vibrating in approximately 
the same period,* 

The fact thus recorded makes several points perfectly 
clear* We learn from it, to begin with, that the blaze of 
hydrogen registering itself in the spectrum occurs quite close 
to the photosphere of Mira, since the stratum of less intensely 
glowing calcium, which absorbs the small quantity of its light 

» Astr. Nach. No. 2924. * Observatory, vol. xi. p. 84. 



it is capable of touching, lies above it The relatively cool 
calcium, however (as the character of its absorption shows) , is 
at just such a pitch of 
thermal or electrical 
excitement as prevails 
in the hottest part of 
the sun's surround- 
ings, while the hydro- 
gen beneath proves 
itself, through its 
emission of the typical 
ultra-violet rays, to 
be at the enormous 
temperature of the 
Sirian atmosphere. 
Conclusive evidence 
seems thus to be 
provided that stellar 
spectra of the third 
type originate at va- 
rious heat -levels in 
powerfully ignited va- 
porous envelopes. 

The dark blue hy- 
drogen line (near G) 
was seen bright in 
Mira by Mr. Maunder 
at Greenwich, October 
5, and by Mr. Espin 
October 23 and 80, 
1888.^ The correspon- 
ding green and red 
rays (F and C) have not 
hitherto been detected. 
Notwithstanding the 
amazing variability of 
this star in the amount of its light-emissions, their kiiid seems 

• Monthly Notices, vol. xlix. pp. 18, 300. 


to remain tolerably constant,* but further inquiries on this 
head are desirable. The bright lines lately discovered by 
Mr. Espin in several objects of the same class come into 
view, it would seem, about the time of maximum general 
brilliancy, and die out as the stars fade. In producing these 
remarkable emanations one other substance is frequently 
associated with hydrogen. This is the enigmatical * helium,' 
known only through spectroscopic observations at the edge 
of the sun, where it occurs in enormous profusion in the 
chromosphere and prominences. Except 'coronium' (the 
chief component of the solar corona), it is the most ethereal 
form of matter we are acquainted with, and in chemical union 
with that still lighter element, it has, with some plausible show 
of reason, been supposed to constitute hydrogen.* The feeble- 
ness of its absorptive action allows its presence in sidereal bodies 
to be signified to us only under the rare conditions occasioning 
the visibiUty of its direct radiations. One of these — perhaps 
the most essential— is a transcendental temperature. The 
evolution of helium lies far beyond the range of terrestrial 
resources for heat-production. Its spectroscopic badge is a 
single line of a deep yellow colour, a little more refrangible 
than the twin-lines of sodium. And since these are known as 
Di and Dj, the helium ray has received the appellation of Dj, 
the progression being towards the blue end of the spectrum. 

Collecting, then, what has been learnt about stars of the 
third type, we find that all are more or less red, that the light 
of all is powerfully absorbed, and that all exhibit highly com- 
plex spectra comprising superposed orders of effects undeniably 
due to very different stages of temperature. It may be added 
that a large proportion are of conspicuously variable bright- 
ness, but the discussion of this peculiarity will be more 
conveniently entered upon in a subsequent chapter. 

These facts suggest the all but certain inference that the 

absorbing atmospheres of these stars are much more extensive 

than that of the sun, and include a wider range of temperatures. 

\The incandescent vapours composing them are, at the photo- 

\5pheric level, as hot as — in some cases much hotter than — 

' Henry Draper Memorial, Third Annual Report ^ p. 5. 
' Griinwald, Aalr, Nach. No. 2797. 


those in a corresponding position in the sun ; but they ascend 
to heights far beyond the limit of the solar mixed metallic 
strata, and are hence competent to produce a cooler kind of 
absorption than any affecting the sun's prismatic light. 

Of this difference we can offer only a speculative explana- 
tion. If it be admitted as probable that the spectroscopic 
transition from Sirian to solar stars represents an enhancement 
of electrical repulsive action producing a partial neutralisation / 
of the power of gravity over their gaseous surroundings, we 
may be permitted to conjecture that the same cause works 
still more energetically in red stars. Their atmospheres are 
inordinately distended because the materials composing them 
are virtually very light ; they are to a great extent emancipated I 
from the thraldom of their own weight ; even the massive 
atoms of bismuth and mercury, buoyed up, as it were, by a 
levitative influence, float at comparatively high levels. The 
flaming appurtenances of these bodies must (should this view 
prove correct) be on a prodigious scale ; it is not impossible 
that visual traces of their abnormal development have been 
perceived. For unless the intermittently hazy aspect of certain 
red stars ^ be due to instrumental causes, it indicates a vast 
diffusion of glowing matter in their neighbourhood. Coronas; 
in fact, hundreds of millions of miles in extent, are on such \ 
occasions made manifest. An approach towards a nebulous I 
condition is thus plainly hinted at. 

In the extreme members of the class, accordingly, the solar 
analogy is weakened to the verge of abolition. The function 
of a * sun ' can hardly be regarded as in any sense discharged 
by such a body as Mira Ceti. But the stage of Mira is only 
reached by insensible gradations. At the other end of the 
scale we meet radiating centres, the efficiency of which is not 
perceptibly compromised by the shaded bands modifying their 
light. As dependents upon An tares (a Scorpii), for instance, 
planets like ours might be warmed and clothed with vegeta- 
tion, and lavishly stocked with life. At the same distances 
they would indeed receive probably several hundred times as 

* See Mr. Peek's observations Knowledge^ vol. xii. p. 126 ; also Pogson, 
Monthly NoticeSy vol. nwi. p. 187. 


much heat and light as from our sun, and they would receive 
it with approximate constancy. Daylight upon them would 
be of a warmer glow than here ; there would be no cold grey 
dawns ; every landscape would be suffused with a flush as of 
autumn ; an universal mellowed illumination would soften 
and harmonise all the features of the scenery. If such phe- 
nomena as total eclipses occurred at all, inconceivably brilliant 
displays of red flames and coronal streamers might, during 
their progress, astonish intelligent spectators, whose observa- 
tions at other times would perhaps disclose the presence of 
spots on their sun, compared with which the sun-spots we 
often wonder at are but puny objects ; but unless their science 
had put them in possession of some method of Ught-analysis, 
they could never know that the body sovereign in their sphere 
was a ' star with a banded spectrum.' 

i I ^ ? ^ ^ ? I ^ liliiiiil 


Fio. 7.- SteUar Spectra of the Fourth Class ; 1, 152 Schjellerap ; 2, 19 
Piscium (Dun6r). 

A comparison of figs. 7 and 4 shows at once the chief 
points of diflference between stellar spectra of the third and of 
the fourth types. The in tenser, bands of absorption in the 
latter, it will be noticed, are only three in number, and they 
face the opposite way from the more frequent shadings in the 
aUied genus. Their sharp sides are towards the red, their 
shaded sides towards the violet end of the spectrum. No 
doubt exists as to their origin ; all three agree accurately in 
position with the bright bands given out by burning alcohol 
and other hydrocarbons. They are the identical cometary 
bands vainly looked for in spectra of the third type, reversed ; 


instead of being prismatically coloured, they are intensely 
dark. The stars showing them might be called * carbon stars ' ; 
they bear, as their discoverer Father Secchi perceived, the 
mimistakable signature of that protean substance which more 
than any other deserves to be called the material basis of life. 
The best authorities agree that the signature is that of pure 
carbon, though there is no practical way of obtaining it except 
through the medium of a hydrogen compound. And since this 
is also very likely to be the case in the stars, it may be as- 
sumed that carburetted hydrogen, or some closely analogous gas, 
forms an important constituent of their atmospheres. The 
absence from their spectra of all traces of free hydrogen is thus 
only what might be expected. Stars of Class IV. are rendered 
the more interesting by their scarcity. About one hundred and 
twenty of them are now known, recent extensions of the list 
being largely due to Mr. Espin's diligent explorations ; and 
be calculates that if as many proportionately are found in 
the southern hemisphere, there may exist in all 168 brighter 
than 8-8 magnitude.* For the most part they are telescopic 
objects. No more than seven or eight can be made out with 
the naked eye. Of these the most brilliant is a star in 
Canes Venatici of 5*5 magnitude, numbered 152 in Schjelle- 
rup's Catalogue of Bed Stars* (fig. 7, No. 1). From the 
extraordinary vivacity of its prismatic rays, it was named by 
Father Secchi ' La Superba,' and in many such stars a mag- 
nificent effect is produced by the sudden alternation of dazzling 

* zones ' of red, yellow, and green light, with spaces of pro- 
found obscurity. Another fine specimen of the class is 19 
Piscium (fig. 7, No. 2), averaging about sixth, but variable to 
the extent of one magnitude. To it also belongs Hind's 

* Crimson Star ' (R Leporis) and many other noted variables. 

The faintness of fourth type stars can be accounted for by 
the extraordinarily powerful atmospheric absorption exercised 
upon their light. The vapours closing in upon them leave 
only an aperture here and there for their rays to escape 
through ; and those more refrangible, as usual, bear the brunt 
of their attacks. The violet end of the spectrum is either cut 

» Monthly Notices^ vol. xlix. p. 364. « Astr. Nach. No. 1591. 


oflF, it might be said, wholly, or greatly enfeebled. Hence a 
6|>ecial depth of colour in these stars. Some appear like ' a 
drop of blood ' freshly fallen upon the field of the telescope ; 
others gleam like clear flame, or glow like carbuncles ; none 
are indifferent or undistinguished in tint. They have accord- 
ingly only of late been rendered accessible to photographic 
investigation, four new specimens having been detected at 
Harvard College through the peculiarities of their self- 
imprinted spectra. 

To the question whether this deficiency of blue rays is to 
any degree innate, or must be altogether attributed to atmo- 
spheric quenching, no categorical reply can at present be made. 
Absorption carried far enough might produce, even in Sirius, 
the observed result ; but it is more likely that imperfect conden- 
sation accompanies strong absorption, and that the photo- 
spheres of 152 Schjellerup and 19 Piscium really fall short, 
in intrinsic brilliancy, of the solar, still more of the Sirian, 
standard. At the same time their powerful incandescence is 
undoubted. Through the obscurity of the cai-bon bands can be 
distinctly seen a * Fraunhofer spectrum,'— a spectrum, that is to 
say, composed, like that of the sun, of dark metalUc lines thrown 
out upon a continuous background. Among the substances 
originating them, sodium is certainly, iron probably, recog- 
nisable ; * but particular identifications are of extreme difficulty 
in objects with so little, and such unequally distributed light. 
Enough, however, is known to assure us that near the photo- 
spheres of fourth type stars metallic vapours are suspended in 
a state of ignition, while the reversal of their characteristic 
rays proves the temperature of the photospheres themselves 
to be still higher. The great depth of their atmospheres is 
certified by the origin within them of two strongly marked 
systems of absorptive effects due to widely different degrees of 
heat ; bright lines of any kind, however, have so far not been 
certainly recorded in these spectra. 

They are, not only in general outline but in particular 
details, of an unvarying pattern ; at the most, some difference 
in the relative intensity of the flutings distinguishes one from 

» Dun6r, Bc^herclies &c. p. 123 ; Vogel, Potsdam Puhlicationen, No. 14, p. 27. 


another. Notwithstanding their obvious affinity to banded 
spectra of the third class, they stand emphatically apart from 
them. Intermediate specimens are completely wanting. 
Between the first and second, and the second and third types, 
they abound ; but the third and fourth are sharply separated. 
No stragglers are picked up between the camps. The objects 
collected in them have nevertheless many qualities in common, 
and among the most striking, that of luminous instability. 
About twelve per cent, of stars with banded spectra are well- 
known variables, the proportion being about equal for each 
kind. Thus, the differences of their constitutions are not sucli 
as materially to affect their inherent tendency to luminous 

Stars with banded spectra are not found indifferently in 
all parts of the sky ; but incline to collect into groups, separate 
for each order. Eight regions of concentration, the principal 
in the constellation Lyra, are indicated for objects of the third 
type ; those of the fourth have their chief gathering-places 
near 7 Cygni and about the tips of the Horns of Taurus.* 
Both these spots lie in the Milky Way, the zone of Which has 
an overwhelming attraction for the ' carbonaceous ' stars. 
All small stars, indeed, are strongly condensed towards it, and 
M. Duner is of opinion ^ that the objects in question merely 
conform to the universal law governing sidereal structure. 
But their obedience to it at least shows them to be situated in 
the region of small, that is, of excessively distant stars, hence 
to be copious radiators, and by no means on the verge of 

• Espin, Observatory, vol. x. p. 259. 

* Spectres de la IIP Classe, p. 12u. 




The fifth spectral class is at present the most restricted of all, 
but its numbers are being rapidly augmented both by photo- 
graphic and by improved visual means. The objects belong- 
ing to it are distinguished by the display in their spectra of 
isolated bright lines on a more or less perfectly continuous 
background, sometimes, however, also interrupted by dark lines 
or bands. They present us then with a triple combination— a 
direct gaseous spectrum, a reversed gaseous spectrum, and a 
spectrum due to glowing solid or liquid matter, all simul- 
taneously made manifest by the unrolling, as it were, of a 
single scroll, yet each originating under very different con- 
ditions. The investigation of what those conditions are con- 
stitutes one of the most important tasks of physical sidereal 

The state of bright emission is in some stars normal, in 
others it only supervenes as part of a great general increase 
of light. This is the case with many * temporary ' and periodi- 
cal stars, their blazing atmospheric constituents being almost 
invariably hydrogen and helium. Objects, on the other hand, 
showing bright lines with approximate constancy, can be dis- 
criminated into two varieties, according as they give or with- 
hold evidence of the presence in them of hydrogen. 

The first specimen of a ' gaseous star ' was made known by 
Father Secchi*s discovery, August 19, 186(), of the green line 
(F) of hydrogen conspicuously bright in 7 Cassiopeice,* the 
middle star of five of the second and third magnitudes grouped 
into the shape of a W on the opposite side of the pole from 

' 8u(jli Spcftri Prismatici dcllc Stcllc Fisse, Mem. i. p. 10, Mem. ii. p. 62. 


the Great Bear. Soon afterwards the same peculiarity re- 
vealed itself in ^ Lyrae, the emission-spectrum of both stars 
consisting of three rays of hydrogen and one of helium. 
But these cannot always be seen, even under the best optical 
cu-cumstances. As early as 1872 Vogel was struck with their 
apparently unaccountable caprices of visibility ; ^ and M. von 
Gothard watched vainly for them during two years before he 
caught sight in 7 CassiopeisB of the crimson twinkling of C 
in particularly unfavourable weather, August 13, 1883.^ His 
subsequent observations, and those of M. von Konkoly,^ fully 
established the occurrence, in both stars, of rapid spectral 
fluctuations, which have recently been studied, with attentive 
curiosity, in widely separated parts of the world. 

The variability disclosed is of a most singular kind. It 
might have appeared a safe forecast that the brightness of the 
lines — of those, at any rate, emitted by the same substance — 
would change in concert, if it changed at all ; experience 
shows, however, that each flares out and dies away, to a great 
extent independently of the others. They take it in turns 
to be brilliant. The stress of variability, however, is in 
7 CassiopeiaB laid upon C,"* in ^ Lyr© upon D3, usually the 
leading feature of this star's gaseous spectrum.* DazzUng at 
times, it often fades to extinction at intervals of a few days ; 
but no definite law has yet been found for its changes. The 
disappearance of the hydrogen-lines is occasionally emphasised 
by reversal. They were seen dark by Von Gothard at Hereny, 
September 5, 1882,^ and probably again by Mr. Maunder at 
Greenwich, October 19, 1888. In 7 CassiopeiaB, the turning 
of the balance appears never to go beyond effacement. Its 
rays do not, however, escape the effects of mixed absorption. 
Mr. Keeler perceived, with the Lick refractor, the green part 
of its spectrum to be * full of very fine, delicate, dark lines, 
seen only under good atmospheric conditions, the h group ' 
(due to magnesium) * being somewhat more prominent than 

' Bothkamp Beobachiungenj Heft ii. p. 29. 
« Astr. Nach. Nos. 2531, 2589. 

* Observatory^ vol. vi. p. 332 ; O Gijalla Bcobachtwigcn vol. viii. p. 6. 

* Copcland, Monthly Notices, vol. xlvil. p. 92. 

^ Maunder, ibid. vol. xlix. p. 300. « Astr. Nach. No. 2581. 

p 2 


the others.' * The sodium-lines at D have been seen dark in 
both stars, but their appearances are temporary. Of a general 
arresting action there is Uttle or no trace. I he white hght 
of these objects remains practically unmodified. 

Stellar bright lines have been explained in several different 
ways. One is by the extent of surface radiating them. Stars 
possessing enormous self-luminous atmospheres would, it has 
been supposed, send us a sum-total of light in which the 
gaseous beams predominate simply through quantitative 
excess over the continuous radiance of comparatively small 
nuclei. And Dr. Scheiner alleges, in support of this view, an 
effect of shading about the sharp, bright hydrogen-rays in 
7 CassiopeisB, due, in his opinion, to absorption by the denser 
strata near the photospheric level.* But it is diflScult to 
beheve in the reality of the state of things thus presented to 
us. In the first place, the fineness of other dark hues in the 
bame spectrum implies an atmosphere tenuous throughout. 
Again, the phenomena of spectral variability appear entirely 
inconsistent with this rationale. Fluctuations in atmospheric 
extent, even if we could admit their occurrence on the in- 
credible scale and with the incredible swiftness required, would 
not account for the relative variability of the bright lines. 
Finally, the extreme complexity lately observed by Dr. Becker 
in the emissive spectrum of /8 LyraB* affords a reasonable 
certainty that it originates in moderately shallow vaporous 
layers. An atmosphere composed solely of hydrogen and 
helium might conceivably rise to vast altitudes, but not one 
of the heterogeneous constitution implied by authentically 
recorded facts. 

It seems, then, as if the display of bright lines must depend 
upon a real excess of atmospheric brilliancy, involving, it 
would seem, the prevalence of a higher temperature in the 
gaseous layers than in the photosphere covered by them. 
Unless, indeed, the same result be brought about by a pecu- 
liarity of electrical condition which we are at present unable 

' Publications Astr. Soc. of tJic Pacific, vol. i. p. 80. 
- SitzungsberichUf Berlin, 1890, viii. p. 8. 
' Monthly Notices, vol. 1. p. 191. 


either to define or to reason about. The spectral variability 
of 7 CassiopeiflB and /8 LyrsB can only arise from thermal or 
electrical instability. Messrs. Frankland and Lockyer showed 
in 1869 that the luminosity of very rare hydrogen can be 
reduced to consist of the green line (F) alone.* The appear- 
ance of C is a symptom of enhanced excitement, so that the 
epochs of its brilliancy in the stars we are considering 
correspond presumably to temperature maxima. The sub- 
stance, on the other hand, of which D, is the spectroscopic 
'device,' is, as the reader is aware, unknown to terrestrial 
chemistry, but thrives in the very hottest part of the solar 
furnace. It is pretty clear, however, that changes of tempera- 
ture will not alone suffice to account for its appearances and 
disappearances. They show no disposition in ^ Lyrae, a 
well-known * short period ' variable, to follow the light-changes 
of the star, tending, indeed, to obey, if any, an inverse law 
connected possibly with the increased effect of contrast as the 
general spectrum becomes enfeebled. 

As to the local origin of such emissions, one item of 
positive evidence is afforded by the Harvard photograph of 
Mira (fig. 6). The stopping-out shown in it by calcium- 
absorption of one line of the otherwise unbroken hydrogen - 
series, gives a perfect assurance that the calcium-vapour lies 
between us and the hydrogen. 

The same state of things doubtless exists in all similar 
stars. That is to say, the bright-line emission associated 
with the light-maxima of variable stars has its seat in a com- . 
paratively shallow layer of vividly incandescent gases at the 
lowest atmospheric level. But it would be unsafe to generalise 
this conclusion too' widely. Miss A. C. Maury's examination 
of some Harvard photographs of the Pleiades showed the 
spectrum of one mamber of the group (Pleione) to contain 
narrow, bright hydrogen-lines superposed upon the dusky 
stripes characterising the Sirian type of stars.* Here, then, 
an absorbing stratum is placed beneath a more vividly incan- 
descent bed of the same substance, and the condition partially 
indicated in 7 Cassiopeise is fully attained. 

» Proc. R. Society, vol. xvii. p. 464. » Pickering, Ast. ^'ach. No. 2934. 


Bright hydrogen, frequently associated with helium and 
other substances (Dr. Becker has laid down seventy bright 
lines in the spectrum of x Cygni), may now be regarded as a 
common feature of the Mira class of variables. It has also 
been detected through photographic researches executed in the 
southern hemisphere by delegates from Harvard College, in 
7j Argus, distinguished for extraordinary light-changes, in 8 
and fi Centauri, as to which there is no suspicion of incon- 
stancy, and, among northern stars, in P Cygni (Flamsteed's 
No. 84 in that constellation). < This last star has a curious 
history. First seen by Janson as a ' new star ' in 1600, the 
revenante of the Swan (as Huygens called it) remained steadily 
of the third magnitude for nineteen years, after which it 
gradually lost light and disappeared. It rose, however, to a 
second maximum in 1659,* and, after some irregular fluctua- 
tions, settled down, towards the close of the seventeenth 
century, at about its present fifth magnitude status. No 
present suspicion of variability attaches to it ; but Bessel ob- 
served it as of 6-7 magnitude, September 14, 1825.^ The 
biographies of other gaseous stars, were they made known, 
might possibly be found to include similar vicissitudes. The 
hydrogen lines in this object, as in Pleione, are doubly 

The finest specimen of the fifth spectral class is invisible 
in these latitudes. First detected by Respighi at Madras, 
December 24, 1871,^ the peculiarities in the light of 7 Argus 
were studied with some care and much delight at the 
* extraordinary beauty ' of the spectacle they present, by Dr. 
Copeland at Puno in the Andes, April 24, 1883.^ An * intensely 
bright line in the blue,' he remarks, ' and the gorgeous group 
of three bright lines in the yellow and orange, render the 
spectrum ' of this star * incomparably the most brilliant and 
striking in the whole heavens.' There is no sign that it is in 
any degree variable. Its appearance to the present writer, at 

*" * Journal des Sgavans, Dec. 166G, p. 288. 

* Gore's Catalogue of Variable Starsy Proc. li. Irish Acad, vol. iv. ser. ii. 
. 199. 

• Comptes Rendust t. Ixxiv. p. 51G. * Copcniicxis^ vol. iii. p. 206. 


the Cape, in October 1888, tallied precisely with Dr. Copeland's 
description, only that the additional feature of a deep band 
of absorption below the cobalt line seemed unmistakable.^ A 
vivid continuous spectrum extends into the violet as far as the 
eye has power to follow it, and accounts for the brilliant 
whiteness of the star. 

Its chemistry, however, remains mysterious. Only 
negative assertions can be safely made about it. Hydrogen 
has certainly nothing to do with producing any of the bright 
lines recorded; and there is great difiSculty in accepting 
Mr. Lockyer's identification of the blue ray with a modified 
carbon-fluting. He may, however, turn out to be right in 
ascribing to sodium-vibrations the faintest and least refrangible 
of the 7 Argus quartette.* Its position, at any rate, lies 
suspiciously near to that of D. 

In the course of his brief explorations from Puno, Dr. 
Copeland came across five small stars of the 7 Argus type ; 
which is also conformed to by two objects discovered by Pro- 
fessor Pickering in 1880-1,' by ten southern stars * spectro- 
graphically' investigated at Harvard College from plates 
exposed at Chosica in Peru, and by a whole group in Cygnus 
(see Appendix, Table I.). The first three of these were 
noted for the singular quality of their light at Paris in 1867 
by MM. Wolf and Kayet,* whose names are hence often used 
to designate this stellar variety ; others were distinguished by 
Dr. Copeland and Professor Pickering. Accurate measure- 
ments of the three original Wolf-Kayet stars, of ' Argelander- 
Oeltzen 17681,' and ' Lalande 13412 ' (so called from their 
numbers in different catalogues), were made with the great 
Vienna refractor by Professor Vogel in 1883, and we have 
his kind permission to reproduce the drawings in which 
he depicted his results (see fig. 8, in which the blue end is to 
the right). In spectrum No. 2, belonging to the eighth 
magnitude star Argelander-Oeltzen 17681, we see the 
prismatic pattern of 7 Argus reduced to its simplest form. 

* Observatory, vol. xi. p. 430. * Proc, R, Society, vol. xliv. p. 33. 

* Nature, vols. xxii. p. 483. xxiii. pp. 338, 604. 

* Comptes Rendus, t. Ixv. p. 292. 


coo wo Md MO *?(> a^ ip^ 

I I I I I f I ir 'I I I I 

4C0 Mm. Mm. 


of Orion may be said to mark the first stage on the road 
towards nebulosity. For their spectra appear at times 
unbroken by the traces of hydrogen-absorption more or less 
strongly impressed upon them at others ; and the transition 
is an easy one from this state of things to that existing in 
B Lyras, where the same sort of fluctuating balance of tempera- 
ture inclines preferentially the other way. That is to say, the 
hydrogen atmosphere of this star tends to rise above the 
thermal level of its photosphere. 'Emissive superiority is 
substituted for neutrality, and gains more and more the upper 
hand as gaseous stars merge into undoubted nebulse. 

In Table I. of the Appendix will be found a Ust of all the 
stars in which bright lines have, up to the present, been 
detected. They are fifty-one in number, and nearly all are 
situated in the very thick of the Milky Way. The only 
gaseous star, indeed, besides a few variables, which can be 
said to lie even shghtly apart from that great collection is 
Muscae. The tendency of these objects to collect in groups 
is very marked, especially among the Wolf-Kayet variety, 
eight specimens of which are congregated in Cygnus, five 
within a restricted area in Scorpio, while two occur about half 
a degree apart, in the Keel of the Ship. Father Secchi's 
conjecture as to the localisation of spectral peculiarities is 
thus proved to have a solid foundation of fact. 

There is only one recognised exception to the rule that the 
lines flashing out in long-period variables are those of hydrogen 
and helium. B Geminorum is a star changing in 871 days 
from above the seventh to below the twelfth magnitude. It 
might be called the * comet-variable,' since the three bright 
bands of carbon usually seen in comets were most probably 
recorded in its spectrum by Vogel as it rose towards a maxi- 
mum on April 7, 1874.* Their display, moreover, was 
combined with that of the typical yellow ray of the Wolf- 
Eayet stars. The significance of the emission by even one 
star in the heavens of distinctively cometary light was 
accentuated by Mr. Lockyer's striking reproduction of its 
range of bright lines in a * meteoritic glow ' at South Ken- 

» Astr, Nach, No. 2000. 


Bington.' Unfortunately, however, no thoroughly satisfactory 
observations of this unique spectrum are extant, for VogeFs 
were made under unfavourable conditions, and have not 
been repeated. They leave, nevertheless, little or no doubt 
that R Geminorum makes no exception to the rule that 
hydrogen and carbon are, in a spectroscopic sense, mutually 
exclusive — that where one appears the other remains im- 

The distinction between gaseous stars and stellar nebulao 
is slight. Both kinds of object present the telescopic appear- 
ance >of small stars, and have, in several instances, been 
registered as such. Both, when a prism is applied, disclose 
analogous peculiarities. Analogous, not identical. The light 
of a gaseous star so examined is ordinarily concentrated in 
two points, or where a cylindrical lens is employed in two short 
lines, yellow and blue respectively; that of a stellar nebula 
gathers into one green knot. Its rays are, in a sense, in- 
capable of analysis; they are so nearly monochromatic that 
they can be refracted without being dispersed. Objects of this 
nature can be picked out at a glance from ordinary stars by 
Professor Pickering's method of * sweeping ' with what wo 
may call a prismatic eye-piece. Above a score of them have 
in this manner been found since 1881 ; * and it is remarkable 
that the exclusive preference for the Milky Way of gaseous 
stars is shared by stellar nebulae. 

The green ray of the latter is the characteristic token 
of gaseous nebulse. From the great Orion * portent ' to the 
faintest ' planetary,' all without exception show it ; and in 
many it is so predominant as virtually to stand alone. But 
its origin remains an enigma. In position it is almost 
coincident with an important line of nitrogen. A trifling 
divergence, however, shows them to be certainly distinct. A 
different view has been urged by Mr. Lockyer. He found 
that fragments of meteorites gently heated in vacuum-tubes 
lit up by the passage of an ele3tric current gave out, pro- 

* Proc, R, Society, vol. xliii. p. 133. 

* Pickering, Observatory^ vols. iv. p. 81, v. p. 29i ; Astr. Nach, No. 2517 ; 
Copeland, Monthly Notices, vol. xlv. p. 91 


minently and persistently, a fluting due to magnesium burning 
at about the temperature of the bunsen-flame. * Now the sharp 
edge of this fluting appeared, with the slight amount of dis- 
persion used in the experiments, to agree sensibly with the 
chief line of nebulsB ; and it was inferred (some other approxi- 
mate ccdncidences pointing, it was thought, towards the same 
conclusion) ' that the nebulae are composed of sparse meteorites, 
the collisions of which bring abopt a rise of temperature 
sufficient to render luminous one of their chief constituents, 
magnesium.' * 

Precise inquiries have nevertheless failed to ratify the 
suggested identification of the fundamental meteoric and 
nebular lines, found by careful measurements to stand very 
slightly apart in the spectrum. But spectroscopic agreements 
must be absolute if they are to be reckoned significant. And 
the disagreement, in the present case, was rendered palpable 
by Dr. and Mrs. Huggins's application to it of the laborious 
method of direct comparison. The injection of the light of 
burning magnesium into the apparatus simultaneously dis- 
playing the spectrum of the Orion nebula had the effect of 
at once producing a well separated doublet. The fluting-edge 
lay markedly lower down towards the red than the nebular 
line.' Nor is it likely that a fluting, however attenuated the 
vapour emitting it, could be reduced to so thin a * remnant ' 
as the sharp, fine line seen in nebulae. 

Nebular radiance cannot then, it seems, be imitated in the 
laboratory. Possibly it signalises a modification of matter 
arising only under extra-terrestrial conditions. The line at 
5005 has, however, been seen in two comets and in one star. 
On January 9, 1866, Dr. Huggins observed the spectrum of 
Comet I., 1866 (of the November meteors), then within two 
days of perihelion, but at a distance from the sun not greatly 
inferior to that of the earth. No trace was seen of the usual 
carbon-bands; the continuous line of prismatic light being 
interrupted only by a vivid point matching the chief nebular 
line in position, and suspected to be accompanied by the 

» Proc. R. Society, vol. xliii. pp. 124, 127. 

» Ibid. vol. xliv. p. 2. » Ibid. vol. xlvi. p. 48. 


second nebular line (at 4957).' Comet II., 1867, apparently 
shared, under closely similar circumstances, the same un- 
common quality.* Again, the *new star' of 1876 shone 
as it faded with a green ray identical with that of stellar 

These bodies, judging by the two illustrative instances just 
mentioned, are not greatly heated. The condensed or nuclear 
portions probably included in them give but slight signs of 
incandescence; spectroscopically (whatever they may be 
physically) they are mere spheres of glimmering gas. Now 
planetary nebute (so called by Sir William Herschel because 
they exhibit a planet-like disc) are not intrinsically different 
from the stellar sort, but they either are, or, owing to their 
greater vicinity, appear larger and brighter, and hence offer 
better facilities for the analysis of their light. 

It is found to consist of three rays composing the funda- 
mental spectrum of all gaseous nebulee. Although of varied 
relative intensity in different individuals, the predominance 
uniformly remains to the line at 6005, which accordingly sur- 
vives alone in such a dearth of light as that created by the 
combined distance and faintness of the monochromatic objects 
detected by Professor Pickering's method. 

The nebular trio of lines (the three to the right in the 
lower part of fig. 1), since they lie adjacent to each other in 
the middle or green part of the spectrum, inevitably give a 
resultant green or bluish colour to the objects they characterise. 
Two of them are unidentified ; the third, or most refrangible, 
is the familiar F line, the never-failing, and frequently solitary 
emanation of the ubiquitous gaseous metal. The undoubted 
and universal presence of hydrogen in nebulae constitutes the 
one link as yet recognised between nebular and terrestrial 
chemistry. But cometary and nebular chemistry are not (as 
Mr. Lockyer first pointed out) wholly disconnected ; and some 
nebular spectra include traces of stellar and solar affinities 
which must prove of the highest importance for determining 
the place in creation of cosmical cloudlets. 

' Proc. R. Society, vol. xv. p. 5. 

• Monthly Notices, vol. xxvii. p. 288. 


Those of the planetary sort display in a few eases,* besides 
the usual three lines, an additional cobalt-blue one most 
likely identical with the blue ray of 7 Argus. It seems also 
to have been very faintly visible, in the spectrum of the 
Orion nebula, to Mr. Albert Taylor, at Sir Henry Thompson's 
observatory at Hurst Side, October 13, 1888.^ But the 
chemical meaning of the line has not been determined. We 
only know that it appears in the majority of gaseous stars, 
and in a very few nebulae and comets. Solar spectroscopy 
ignores it. 

Although the spectrum, both visual and photographic, of 
the last-named marvellous object has been diligently studied, 
our knowledge of it is still far from complete. It includes, over 
and above the normal three, a considerable number of lines, 
the relationships and conditions of appearance of which oflfer 
a tempting subject of investigation. In other nebulae hydrogen 
attests its presence solely by its green ray. But F is here 
attended by G, h, H, and the two first of the ultra-violet 
series,^ while the crimson C line remains imperceptible. 
The hydrogen, then, entering into the composition of the Orion 
nebula is at a different, perhaps a higher, stage of excitement 
than in the sun itself. It is associated with luminous helium. 
The D3 line detected by Dr. Copeland, December 28, 1886, and 
subsequently measured by him thirty times,"* had its identity 
confirmed by Mr. Taylor's observations. 

The diffused Ught about the central group of rays in this 

object shows. Dr. Coi)eland remarks, 'some indications of 

resolvability into lines or bands ; ' and Dr. and Mrs. Huggins 

express the opinion that the faint continuous spectrum visible 

in most gaseous nebulae might, were more light available, * be 

found to consist, in great part at least, of closely adjacent 

bright lines.' ^ 

' Copeland, Copernicus, vol. i. p. 2. The * four-line' planctarics are 
numbered in Dreyer's New General Catalogue, 7602, 7026, and 7027. The two 
last belong to the singular group of allied gaseous objects in Cjgnus. 

* Monthly Notices, vol. xlix. p. 121. 

» The lines h and H were photographed by Mr. Lockyer early in 1890, tlie 
entire scries a little later by Dr. and Mrs. Huggins. 

* Monthly Notices, vol. xlviii. p. 360. 

* iVoc. li. Society, vol. xlvi. p. 60. 



Nearly all that is known as to the photographic spectrum 
of the Orion nebula has been elicited by the same eminent 
inquirers, and many features in their results are of extreme 
interest. A plate exposed in 1882 contained a strong ultra- 
violet ray of about wave-length 8725 (see fig. 1) which reap- 
peared February 6, 1888, with a number of fainter companions 
as shown in the upper section of fig. 9. Now the light admitted 
on this occasion to the camera was obtained from the vivid 
central mass of the nebula, the slit in fact lying across two of 
the four bright stars grouped into the 'trapezium,' which 
forms apparently the core of the entire formation. The pair 
of stellar spectra thus included in the photograph are seen in 
the figure as two unbroken strips of light, crossed by the 






Fig. 9.— Photographic Spectrum of tho Great Nebula in Orion (Huggins). 

gaseous rays of the nebula. But not of the nebula alone. 
Three groups of thin and feeble lines belong, so far as it is 
possible to judge, primarily to the stars, though they are also 
emitted by the nebulous stuff in their immediate neighbour- 
hood. The stars then, since an indefinite extent of circum- 
jacent vaporous material partakes of the nature of their 
glowing atmosf^heres, are in the nebula, and not merely pro- 
jected upoji it.* They form an integral part of its structure. 

Their gaseous nature was unsuspected until disclosed by 
this negative. No bright lines had been seen in their spectra. 
It remains to be decided whether the trapezium stars are to 
be regarded as sui c/cneris, or whether they will prove to bo 
the first known members of a new class. Several coincidences, 

' i'7V)r. It. ScKietj/y vol. xlvi. p. 41. 


some approximate, some exact, between the ultra-violet rays 
emitted by them and those absorbed in the atmospheres of 
Are turns and a Aquilae (as photographed by Dr. Huggins in 
1879) are apparent, and may serve as an index to their 
physical condition.* They, at any rate, suggest a practicable 
path for further investigation. 

The next successful photograph of the nebular spectrum 
was taken a year later than its predecessor, February 28, 
1889. Its curious unlikeness to it can be seen in fig. 9 (lower 
part), where the spectrum of 1888 survives only in the 
subordinate pair of lines on the extreme right (identical with 
those on the extreme left above) ; even the predominant ultra- 
violet ray at 3725 has vanished. The change can only be 
attributed (as was pointed out by the Tulse Hill astronomers) 
to local diflFerences in the constitution of the nebula. For the 
slit lay in the second experiment at some distance from the 
trapezium, and the light it transmitted originated consequently 
from a less condensed part of the structure than that ex- 
amined on the previous occasion. This variation of the 
photographic spectrum is the more surprising from the funda- 
mental sameness of the visible spectrum of the nebula. 

We learn from it the important lesson that the spectra of 
the great nebulae, like those of the * zoned ' stars, must be 
considered as integrating the results of emanations taking 
their rise under notably diverse circumstances. Innumerable 
strata of nebulous matter are piled one upon the other in the 
same line of sight. The eye is impotent to discriminate be- 
tween them ; even the spectroscope can do so only indirectly. 
For, at the centre of the nebula, the lines coming from all its 
depths are seen or photographed together; their respective 
local origins are unnoticed. Light, on the other hand, taken 
from near the edges of the same object, emanates exclusively 
from its higher regions, and its characteristic peculiarities 
may safely be referred to that circumstance. The possibility, 
then, seems at hand of dividing in this way the Orion nebula 
and others of the same class into various spectroscopic levels, 
distinguished by minor emissive differences. The l)^ lino, for 

' Observalonj, vol. xii. p. 367. 


instance, may prove separable from the 7 Argus blue line, and , 
this again from Dr. Copeland's line at 4476. As to the fun- 
damental trio, they are doubtless common to all the heights 
and depths, nodes and convolutions, of gaseous nebulae. The 
chemical composition of the much scrutinised one in Orion, 
so far as we yet know it, agrees, we cannot fail to remark, with 
that of the solar prominences; a result partial indeed, yet 
tolerably secure, even if somewhat diflBcult to reconcile with 
any hitherto admitted theory. 
^ Only the species designated as stellar, planetary, annular, 
and irregular nebulae, give unmistakable signs of gaseity. The 
rest, and they are the great majority, shine with continuous 
light; yet the distinction is perhaps less profound than it 
seems. The gap is at any rate partially bridged. One • fron- 
tier instance ' seems to be supplied by the great * looped 
nebula ' in the southern constellation of Dorado, observed by 
Mr. C. E. Burton* in 1874 to yield a strongly continuous 
spectrum crossed by the unfailing green nebular ray at 5005, 
Its gaseous nature is thus shown to be modified by the pre- 
sence of an unusually large proportion of dense material, and 
where this predominates, as in the Andromeda nebula, the 
spectrum, though nominally 'continuous,' is still markedly 
different from the continuous spectrum of a star. 

The light of this ' queen of the nebulae,' prismatically dis- 
persed for the first time by Dr. Huggins in 1864, was found 
to end abruptly in the orange ; it included no red rays, nor 
was it anywhere uniform, but seemed mottled throughout, 
whether by the effects of absorption or of irregular emission, 
it was impossible to decide.^ Recent observations leave the 
point still unsettled, the existence of various alleged bright 
lines or spaces needing further confirmation. In the investi- 
gation, nevertheless, of the distinctive peculiarities of * con- 
tinuous ' nebular light, a field lies open which can hardlj'^ fail 
to be worked with profit to the rapidly advancing science of 
cosmical physics. 

* Monthly Notices, vol. xxxvi. p. 69. * PhiL Trans, vol. cliv. p. 4-11. 




Are the stars subject to growth and decay? We might 
almost as well ask, Are they subject to the laws of nature ? 
There can be in either case no doubt about the reply. We 
are perfectly assured, both from reason and revelation, that a 
time was when they were not, and that at some future date 
they will have ceased to be. And we may further confidently 
affirm, guided by the analogy of all other creative processes 
with which we are acquainted, that their present condition 
has been gradually attained and will gradually become 

Each has then a life-history. It is what it is, because it 
has been what it was. Nor is it conceivable that all should 
have arrived simultaneously at the same stage of develop- 
ment. A contemporaneous universal origin can by no means 
be assumed as a postulate ; and even if it could, the rate of 
progress of individual stars must have been indefinitely 
varied. There is hence a strong probability that the present 
state of some represents the past of others, the future of many 
more. Among the hosts of heaven we may expect to find 
stars in embryo, stars half formed yet chaotic, full-grown 
stars in orderly and equable working order, stars still effective 
as radiators though of declining powers, and stars on the 
verge of decrepitude. Their comparative study ought then, 
under certain conditions, to enable us to compile, as it were, 
the typical biography of an average star. 

This is a grand idea, but one not easy of legitimate reali- 
sation. The criteria of stellar * age * or * youth ' are far from 
obvious, and hasty conclusions about them are more likely 
than not to prove illusory. There are certain principles, 


however, which we can hardly be mistaken in adopting as our 
guides. In all cosmical masses, a perennial contest is in pro- 
gress between heat and gravity. Heat strives for expansion, 
gravity for contraction. And gravity must in all cases win in 
the end, for this simple reason that, whereas heat is conti- 
nually wasted by radiation, gravity remains unalterably the 
same. Its action is (for what reason we cannot tell) without 
expenditure. Condensation then attends upon the efflux of 
time, and its degree, other things being .equal, measures 
antiquity. One of its effects is to renew, up to a certain point, 
the thermal stores upon the gradual dissipation of which it 
ensues. For condensation implies a fall of each particle of 
the condensing body towards its centre, and all their arrested 
motions turn into heat as inevitably as percussion evokes sparks 
from a flint. Moreover, the production of heat by contraction 
exceeds the loss by radiation so long as the body in question 
remains purely gaseous, but begins to fall short of it with the 
approach of liquefaction. A gaseous mass condensing in 
space under the influence of its own internal forces will thus 
rise progressively in temperature until a change of its state is 
imminent, radiation going on for ages without cooling, because 
the loss of heat is more than compensated by the attendant 
transformation of gravitational energy. 

The course of development of a star leads it then slowly 
but surely from a rare to a denser condition, its volume per- 
petually diminishing as it advances in life. Its changes of 
temperature obey a more complex law. They at first, and 
during countless millions of years, take an upward direction ; 
at last a maximum is reached, the tide turnsj and a decline 
sets in, unintermitted until the former sun, reduced to an 
arid rock, has sunk to the level of the unimaginable cold of 
space. It must not be forgotten, however, that all this, 
though true, may not be the whole truth. There are forces in 
nature powerfully influential, doubtless, in determining phases 
of development, as to the working of which we are almost wholly 
ignorant. What do we know, for instance, of the electrical 
state of gaseous nebulse ? That their shining depends upon 
it is forcibly suggested by the quality of their light ; but as to 

o 2 


the conditions prescribing electrical changes in cosmical bodies, 
we are without a clue. Only their effects can here and there 
be hesitatingly traced. 

We must accordingly be prepared, in examining the 
heavens, to meet with apparent anomalies. All our specimens 
will not suit the labels we have prepared for them. Nor can 
it be supposed that all the stars are made on one pattern, and 
follow each other along the same rigid groove of change. 
There are to be found not only many different species of the 
heavenly bodies, but almost endless varieties of the same 
species. The universe is no mechanical workshop, turning 
out objects by the score in blind pursuance of one original 
intelligent arrangement, but a scene of continuous and ex- 
quisite adaptation of means to the most subtly various ends. 

The task of settling the mutual relationships of the sidereal 
classes, ranging them in orderly sequence, and interpreting 
historically the physical links that unite them, is an arduous 
one. But at least on one point there is unanimity of opinion. 
The nebulous state preceded the stellar. The generation of 
stars from nebulse was inferred by Sir William Ilerschel in 
1811 and 1814 as the upshot of * a critical examination of the 
nebulous system,' showing extreme instances from each of the 
sidereal kingdoms to be * connected by such nearly allied 
intermediate steps as will make it highly probable that every 
succeeding state of the nebulous matter is the result of the 
action of gravitation upon it while in a foregoing one.' * This 
conclusion, based upon purely telescopic evidence, has been 
fully ratified by the spectroscope. In quality of light, as well 
as in general aspect, distinctions between stars and nebulae 
are shaded off by numerous minute gradations. There is no 
real breach of continuity anywhere. The line-spectra of one 
division of nebulae include continuous radiance ; the continuous 
spectra of the other division possibly include bright lines. 
Gaseous stars take their rise almost insensibly from planetary 
nebula), and themselves merge into unmistakable suns. That 
the great nebulae are the parent forms of stellar clusters is 
rendered highly probable by their common possession of 

» rhil Tr^ns. vol. ci. p. 330. 


luminous as well as structural peculiarities ; nor can any 
definitive separation between them be established. The tra- 
pezium-stars in Orion, like crystals embedded in their rocky 
matrix, are still thickly folded in the generating cosmical stufif. 
By Dr. Huggins's photograph, they may be said to be ' caught in 
the act ' of completing their transformation, a partial survival of 
the original community of gaseous nature being made apparent 
through their self-recorded bright lines. In the stars of the 
Pleiades a further stage of advance is exemplified. Nebulous 
appendages are, in their case, reduced to a subordinate 
position ; the group is essentially a collection of most eflfulgont 

But when we come to the various classes of stars, the 
order of their succession is less easily determined. The 
earliest and most obvious idea on the subject was based on a 
false analogy between the colours of the stars and the colours 
of glowing terrestrial solids. Red stars, it was thought, 
should be regarded, because they had cooled from a condition 
of white heat, as older than white stars.' The colours of 
stars, however, depend j)rimarily upon the quality and extent 
of their absorbing atmospheres, and quite secondarily upon 
their stage of incandescence. A more fruitful suggestion was 
made by Angstrom in 1868.^ It was that of seeking a test 
both of age and temperature in the chemical composition of 
stellar atmospheres. Mr. Lockyer acted upon it in his 
Bakerian Lecture in 1873,' and it formed the basis of Vogel's 
scheme of classification published in 1874.* Here the ances- 
tral type of all other stars was found in those of the Sirian 
pattern of spectrum. Our sun was assumed to be a some- 
what decayed and declining Vega ; Betelgeux to represent a 
still further deteriorated luminary. The two varieties of 
banded spectra, however, marked two alternative routes to 
extinction. Between them a choice had, at some certain epoch, 
to be made, since no transition from one to the other was 
admitted. Hie locus est, partes uhi se viajindit in amhas. 

' ZdllneFf PhotometriscJie Untcrsuchungen^ p. 243. 

' Recherches sur le Spectre Solaire^ p. 38. 

• Phil. Trans, vol. clxiv. p. 492. * Astr. Nach. No. 2000. 


The objection, however, could not fail to present itself that, 
in this arrangement, growing suns had no place.^ Ignoring 
the rising branch of the temperature-curve traversed by all 
condensing and radiating bodies, it regarded only the latter 
half of stellar careers. From the acme of splendour it traced 
them downward to the dimness of impending incrustation, 
but gave no account of the stages by which the acme of splen- 
dour was reached. Yet the nebulous affinities, of late photo- 
graphically ascertained, of certain white stars, give so strong, 
if only a partial support to Vogel's scheme, as to preclude its 
unqualified rejection. What had seemed its essential deficiency 
was supplied by a totally different arrangement proposed by 
Mr. Lockyer in the Bakerian Lecture of 1888.* 

This remarkable work might be described as a project of 
unification of all the * celestial species * — suns, planets, and 
comets, stars * of sorts,' and nebulae. All are exhibited in it 
as the outcome of the aggregation in space of stony and 
metallic fragments, some specimens of which the earth yearly 
intercepts and captures in its orbital course, and deUvers over 
to the inquisitive scrutiny of savants. Hence the possibility 
of producing such evidence as that with which Mr. Lockyer 
supported his views. It consisted in a number of apparent 
spectroscopic coincidences between characteristic stellar and 
nebular rays, brought out by causing vaporised meteoritic par- 
ticles to glow by electricity in the laboratory. But the coin- 
cidences professed to be no more than approximate, and were 
accordingly admissible as suggestions of great interest, but 
not as demonstrations. 

The general conclusions arrived at may be thus sum- 
marised.^ All self-luminous bodies in space are composed of 
meteorites variously aggregated and at various stages of tem- 
perature depending upon the frequency and violence of their 
mutual colUsions. Comets, nebulae, gaseous stars, and stars 
showing banded spectra of the third type, are veritable meteor- 
swarms — they are made up, that is to say, of an indefinite 
multitude of separate, and, in a sense, independent solid 

' Nature, vol. xxxiii. p. 585. ' Proc. R. Society, vol. xliv. p. 1. 

■ Observatory, voi. xi. p. 84. 


bodies, bathed iu evolved gases, and glowing with the heat 
due to their arrested motions. Their component meteorites, 
however, eventually become completely vaporised, when stars 
of the solar and Sirian types are produced. These — the only 
true ' suns '—owe their high temperatures to the surrendered 
velocities of the original myriads of jostling particles drawn 
together to constitute them by the victorious power of gravity. 

To the meteor-swarms thus serving as the basis of stellar 
transformations, a collective origin, and a collective history, 
are necessarily attributed. The theory before us postulates 
a 'curdling process,' by which from some unimaginably subtle 
kind of matter, an ' infinitely fine ' metallic dust is formed » 
and this infinitely fine dust ' becomes at last, in the celestial 
spaces, agglomerated into meteoric irons and stones.' ^ The 
* eddying' of these round self-constituted centres gives the 
meteor-swarms supposed to represent the protoplasm of stars. 

But their members, so far as the specimens ranged on the 
shelves of our museums enable us to judge, by no means 
display the primitive character which, on this hypothesis, they 
ought to possess. They are, on the contrary, minerals of 
highly complex constitution, the products, apparently, of 
lengthy and intricate processes such as have been going on 
for ages in the bowels of the earth.^ Among them are found 
(with certain minor, yet characteristic differences) genuine 
breccias, serpentines, and lavas ; a diamond-bearing meteorite 
recently fell in Siberia ; ^ while in the Deesa meteorite we 
have a splinter from a vein of iron injected, it would appear, 
into a previously existing rock on some unknown planetary 
globe.* The geology illustrated by these samples from distant 
rock-factories, so strangely dropped out of the clouds at our 
feet, belongs indeed to a pre-aqueous epoch ; * they were pro- 
bably formed in the presence of free hydrogen. But we need 
not here stop to discuss the obscure question of their origin. 
What immediately concerns us is the strong presumption 

1 Nineteenth Century, Nov. 1889, p. 787. 

' Stanislas Meanier, Comptes Rendus^ t. cv. pp. 1038, 1095 ; t. cvii. p. 834. 

» Ibid. t. cvi. p. 1678. 

* Meunier, Giologic Cotnparie, p. 83. 
» F61ix H6ment, Les Etoiles Filanies ct les Mit4orites, p. 96, 1888. 


afforded by their structure that they are either debris or 
ejecta, and were not separately formed in space. Nor can 
processes of vaporisation through mutual impacts followed by 
re-crystallisations, be brought in to explain the complexities 
of their constitution. For this reason, if for no other, that 
since heat is only produced at the expense of motion, vapours 
expelled as the result of collisions could not condense into 
circulating members of the system, but would slowly fall to 
its centre. Individual meteorites, which by the hypothesis 
subsist only by virtue of their circulatory movements, could 
not then be formed by them. 

Bodies of this class doubtless form a link in the cosmical 
chain, and are united by relations of great consequence with 
larger cosmical masses, but data are still wanting for the 
satisfactory determination of what those precise relations are. 
Spectroscopic agreements are not decisive ; to a certain extent 
they are, indeed, inevitable, through the fundamental unity 
underlying (there is reason to suppose) the chemistry of all the 
celestial species. About thirty terrestrial elements, for instance, 
occur in meteorites ; and many, if not all of these, are, we 
may be sure, in one way or another universally diffused in 

Mr. Lockyer's classification of the heavenly bodies may, 
however, be considered independently of his meteoric hypo- 
thesis. To a great extent it stands on its own footing ; and 
it well deserves thinking about. Essentially an evolutionary 
scheme, it is the first of the kind with any pretension to com- 
pleteness. Stellar destinies are traced in it, so to speak, from 
the cradle to the grave. The whole history is placed before 
us, of how, by the ceaseless advance of condensation, nebulsB 
are transformed, first into gaseous stars, then into stars with 
banded spectra of the type of Betelgeux, from which solar 
stars, and from these again Sirian stars, gradually emerge. 
Here the ascent ends ; the maximum of temperature is 
reached, and a descent begins, the initial stage of which is 
marked by a second group of objects like our sun and Capella, 
distinguished from the first by the circumstance that they are 
losing, instead of gaining heat ; while low^er still the condition 


immediately antecedent to solidification and obscurity is re- 
presented by Father Secchi's * carbon-stars.' 

The notion that red stars are more aged— nearer to their 
end as suns — than white, has little to recommend it. Mr. 
Lockyer's principle that condensation measures age is really 
fatal to it. For stars with banded spectra, of both kinds, are 
almost demonstrably less condensed bodies than stars of 
which the light is interrupted by linear absorption. Those of 
the third and fourth types alike possess tx)mplex spectra 
originating at various heat-levels in glowing atmospheres, 
which permit the escape of but little of their photospheric 
radiance^ and that chiefly of the less refrangible quahties. 
They are red, not necessarily in themselves, but because we 
see them as if through a shade of tinted glass. Each type of 
red star moreover furnishes examples of the intermittently 
hazy appearance earlier alluded to ; and each shows an equal 
tendency to wide fluctuations of light, connected by a rea- 
sonable surmise with the vast compass of their gaseous 
surroundings. They cannot then, with any probability, be 
set far apart in the developmental series. The place near its 
close assigned by Mr. Lockyer to fourth-type stars, is indeed 
avowedly only provisional. 

There are many indications that these objects are not only 
in an elementary phase of growth, but of exceptional 
character. The peculiarities of their distribution separate 
them from ordinary stars ; their spectroscopic isolation leaves 
us without the means of tracing their relationships. With 
comets, and with at least one gaseous star, they have indeed 
a close affinity. Thus the chief distinction between 19 
Piscium and B Geminorum would be abolished by heighten- 
ing the atmospheric incandescence of the one, or the photo- 
spheric incandescence of the other body, so as to turn dark 
bands to bright in the first case, bright bands to dark in the 

But if the fourth stellar type must for the present be 
looked upon as a sort of cul de sac, a main road obviously 
puts the third in communication with the second type, and 
the second with the first. The transition from each to the 


next is by such easy steps that it is impossible to say 
precisely where the boundary is crossed. They are moreover 
invariably of the same tenour. As the fluted shadings 
exemplified by Betelgeux die out, they uniformly leave behind 
them a line-spectrum of the solar, never one of the Sirian 
pattern. Solar stars then exhibit the transition from such 
stars as Betelgeux to such stars as Sirius and Vega. Their 
intermediate position is unmistakable, and has long been 
recognised. The only difference of opinion is as to the 
direction of progress. 

On this point the consideration of relative atmospheric 
extent may serve as a guide. The vaporous envelopes of 
glowing globes almost necessarily become less voluminous 
with time. As cooling progresses they slowly subside, and 
eventually all but disappear. Our own earth, when its 
temperature was above 212°, kept its oceans in the form of 
dry water-gas suspended above its surface; and that the 
steadily advancing process of atmospheric attenuation has 
not reached its term, we have only to look at the moon to feel 
assured. Star spectra slightly impressed by absorption ought 
then (so far as can be determined from this single point of 
view) to rank as posterior to those strongly so impressed. 

This involves the question of the sun's standing. How 
far has its development proceeded ? What will be its next 
stage? What place should be assigned to it in the grand 
temporal procession of the skies? Categorical answers to 
these questions cannot, of course, be given; but we can 
surmise that at some indefinitely remote epoch the sun was a 
red star with a banded spectrum like that of Betelgeux ; that 
after many ages it came to resemble Aldebaran ; and that as 
the veil thrown over its violet light grew thinner, its present 
mellowed splendour was at last attained. But there is a future 
to be considered as well as the past. The solar transformations 
are certainly not terminated. Will they include a * white 
star ' phase ? It seems possible that they may, since the 
intense electrical excitement by which the sun's ecUpse-appen- 
dages are now maintained will presumably relax ; with results 
perhaps in the subsidence of metallic vapours, and the con- 


centration of hydrogen-strata, issuing in the production of 
the characteristic spectrum of Vega or of Castor. 

We know of no reason why the sun should not be growing 
hotter instead of cooler. There is not a shadow of proof 
that the earth received from it, in past geological ages, 
more heat and light than it does now ; and whatever changes 
may be in progress are beyond doubt immeasurably too slow 
ior detection within the historical period, Mr. Lockyer, it is 
true, feels the difficulty to be insuperable of attributing to the 
sun a prospective condition analogous to that now prevailing 
in Sirius; and he accordingly places our luminary with 
Gapella, Arcturus, and some other selected solar stars, on the 
descending branch of his curve, while a corresponding post on 
the ascending branch is occupied by the rest of their compeers. 
It is, however, scarcely conceivable that a state aboUshed as 
an effect of condensation should be restored by its further 

Under what spectral category then are stars past their 
prime to be disposed ? The phenomena of double stars may 
help us to a reply. The comparative development of members 
of binary systems was first systematically considered by Mr. 
Lockyer in a suggestive inquiry as to their origin.^ Since 
the capture of one star by another is virtually impossible, it 
may be assumed that coupled objects are products of a single 
nebula, that they are related by especially close affinities, ' ^'i 
physical and chemical, and that they have grown up side by S ^'^ »y 
side under identical conditions. Inequality of mass consti- v.}- 

tutes the one certain difference between them. But inequality »* ^ '^ 
of mass involves an unequal rate of development, since a f v 

small star must, it would seem, run through its changes more ' \ 
rapidly than a large one. We have only then to compare the 
spectrum of a satellite-star with that of its primary in order 
to learn the chronological succession of the types they re- 
spectively present. But this presupposes a knowledge of their 
relative masses far from easy to be 'obtained. Light measure- 
ments are indecisive ; the less massive may, during asons of 
time, be the brighter object ; and other methods are usually 
* Proc. B, Sccietyy vol. xlv. p. 250. 


arduous, if not impracticable. In only one revolving system, 
indeed, has mass been as yet satisfactorily apportioned. It 
is that of Sirius, the faint companion of which contains just 
half its amount of matter. Here then we have a star which 
ought to be running down hill, and which, judging from its 
extremely feeble luminosity, actually has advanced far towards 

The spectrum of the Sirian satellite may hence with con-, 
fidence be taken as typical of the spectra of waning stars. 
But here again, observations are wanting, nor can they bo 
procured during the present close conjunction of the pair. 
There is a strong probability, however, that the light of the dim 
component will prove, on analysis, to be of an undistinguished 
character, interrupted neither by bands nor conspicuous dark 
lines, and feeble, not through eflfects of absorption, but intrin- 
sically. The same dull uniformity may be expected to belong 
to the spectra of all stars of impaired splendour ; but on this 
subject we shall know more when some progress has been 
made in determining the relative masses and 8i)ectra of binary 

To resume. The line of stellar evolution indicated by 
recent inquiries is from red stars with banded spectra through 
yellow stars with metallic Une-spectra to white stars distin- 
guished by almost exclusive hydrogen-absorption. In these 
the accentuation of the photosphere as a radiating surface 
culminates, and its temperature reaches a maximum. Even- 
tually, however, the heat of contraction begins to fall short 
of the heat sent abroad through space, supply ceases to 
cover expenditure, and definitive cooling sets in. Stars in 
this condition probably emit light especially feeble in the 
upper prismatic reaches, and give spectra of scarcely deter- 
minable type. They may plausibly be regarded as doomed 
to complete extinction. 

EeturnLng to the opposite end of the scale of stellar being, 
we find its elements universally in nebulae of one kind or 
another. Yet the steps by which stars with banded spectra 
have taken form out of a primitive diffuse substance, remain 
largely hypothetical. On the other hand, certain groups of 


-white stars appear to have arisen directly from parent nebute, 
traversing probably no preliminary phases save that marked 
by a bright line spectrum. Such are the stars in Orion and 
the Pleiades. The latter, though evidently more fully de- 
veloped than many of the former, have scarcely yet, if we may 
say so, been turned out of the workshop, where scraps and 
shavings of the material used in their construction cling to 
them and strew surrounding space. These singular relations 
are not, it must be acknowledged, easily reconciled with the 
ideas we had, on other grounds, been led to form as to the 
status of red stars. They warn us to suspend our conclusions. 
In the mean time, it is permissible to remark that the whole 
stellar army are not bound to advance along the same road, 
their individual constitution includes much that is inscru- 
table to us, nor can it be identical for all. 

We should expect to find the prevalence of the several 
stellar types a rough measure of their durability, and their 
durability determined by the emissive intensity characterising 
them. The longest stage ought, on this view, to be the most 
numerously represented, and that stage ought to be the longest 
in which cooling and the changes consequent upon cooling 
make the slowest progress. But this anticipation is not 
realised. The red stars, although their state should be rela- 
tively permanent owing to the feebleness of their radiation, 
are few compared with Sirian stars, the reckless profusion of 
which in the dissipation of energy seems premonitory of an 
imminent decline ; while solar stars occupy in both respects 
an intermediate position. A rapidly increasing series is in 
fact substituted for the looked for rapidly diminishing one. 
The inference suggested is that the numbers of objects simul- 
taneously belonging to the various spectral classes is a 
question of epoch. The world is not now as it always was. 
There has been a beginning, there will be an end. Why 
should we not admit that between the beginning and the end 
progress may be measured by definable landmarks? We 
have reached a time when the majority of the stars are at 
their highest lustre ; there may have been a time when they 
were mostly red and variable, and there will perhaps be a 


time when only the red stars of to-day still shine inihe sky, 
and the Sirian stars of to-day have sunk into obscurity. 

For the best part of a century the nebular hypothesis, 
stellar and planetary, satisfied scientific curiosity as to cos- 
mical origins. It has, however, ceased to do so. The con- 
sciousness has become more and more insistent that it is not 
enough to refer stars to nebulse, while nebulae themselves 
remain unaccounted for. The need presses ' to explain our 
explanation.' * Pre-nebular ' theories have accordingly come 
to be of the order of the day. One such is afforded by Mr. 
Lockyer's meteoric views; another, due to Dr. Croll,* supposes 
space primitively occupied by cold dark masses moving with 
great but random velocities, and producing very hot nebulae 
by their colUsions. But it may be remarked, in the fij-st 
place, that the odds against the occurrence of any such col- 
lisions are Uterally 'beyond arithmetic*; in the next, that 
the requisite stock of heat might just as well be created 
directly in the form of molecular motion, as indirectly in the 
form of molar motion. Such efforts to get nearer to an abso- 
lute beginning illustrate the incapacity of the human mind to 
rest finally in any purely material conception. It wanders 
vainly from theory to theory, piling Ossa upon Pelion in the 
shape of hypotheses, vainly hoping that, with the help of the 
last and latest, the empyrean may at length be scaled. Har- 
mony can only be established between its aspirations and the 
outer show of the world, and science can only become truly 
rational, when the fount of all things is reached in an Intelli- 
gence akin to, yet infinitely transcending its own. 

' Stellar Evolution and its Relations to Geological Time, London : 1889. 




The facts connected mth the light-clianges of stars are in the 
highest degree strange and surprising ; and wonder is not 
lessened by our daily-growing familiarity with them. They 
are of everyday occurrence, they can be predicted beforehand, 
in many cases with nearly as close accuracy as an eclipse of 
the sun or moon, and they affect in manifold wa,>s a great 
number of objects. Stellar variability is of every kind and 
degree. With the regularity of clockwork some stars lose and 
regain a fixed proportion o^ their light ; others show fitful 
accessions of luminosity succeeded by equally fitful relapses 
into obscurity ; many waver, in appearance lawlessly, about a 
datum-level of lustre itself perhaps slowly rising or sinking. 
The rule of change of a great number is that of an evident, 
though strongly disturbed periodicity ; a few seem to spend 
all their powers of shining in one amazing outburst, after which 
they return to their pristine invisibility or insignificance. 

The amount is as much diversified as the manner of fluc- 
tuation. Changes of brightness so minute as almost to defy 
detection are linked on by a succession of graduated examples 
to conflagrations in which emissive intensity is multiplied a 
thousand times or more in a few hours. The range of varia- 
tion is in some stars sensibly uniform ; they subside during 
each crisis of change to the same precise point of dimness, 
and recover, without diminution or excess, just so much light 
as they had before. In others it is widely irregular. The 
limits of fluctuation in one period furnish no precedent to be 
conformed to in the next. Nothing is predetermined; the 
intensity of each phase seems to depend upon a complex set 


of conditions unlikely to recur twice in the same precise com- 

The first eflfort to rationalise the phenomena of variable 
stars was made by Professor E. C. Pickering in 1880.* His 
five classes, though often enough (as might be expected) con- 
fused at the borders, are still sufficiently distinct to form a 
useful framework for the facts. They are as follows : Class I. 
includes temporary or * new ' stars ; Class II., stars like * Mira ' 
Ceti, strikingly variable in periods of several months ; Class 
III., stars showing slight and irregular fluctuations ; Class IV., 
variables with periods of a few days exemplified by S Cephei 
and /9 Lyrte ; Class V., * Algol-variables,' or stars resembling 
Algol in Perseus in rapidly losing and regaining a determi- 
nate amount of light at intervals measured by hours. We 
will take each in turn, beginning with the first. 

A temporary star may be defined as a variable attaining 
one brief, vivid maximum. An extraordinarily swift rise to 
such an extent as to constitute a virtually * new ' object, 
followed by a slower yet prompt decline, characterise these 
outbursts, above a score of which have been more or less 
credibly recorded within historical times. Those connected 
with the following years * are some of them probably, most of 
them certainly, genuine. 

184 B.C. in Scorpio ; the star of Hipparchus. 

123 A.D. in Ophiuchus. 

Dec. 10, 173, between a and fi Centauri. Conspicuous ; 
scintillated strongly ; visible eight months. 

386 (April to July), between X and ^ Sagittarii. 

389, near a Aquilse, said by Cuspinianus to have equaUed 
Venus ; vanished after three weeks. 

March 893, in the Tail of the Scorpion. 

827 (?), in Scorpio. Observed during four months at 
Babylon. There is some uncertainty about the date, none 
about the fact. 

May 1012, in Aries. Described by Epidamnus, the monk 
of St. Gall, as * oculos verberans.' 

> Proceedings Amer. Acad. yol. xvi. p. 17. 

s See Humboldt's Cosmos, yol. iii. p. 209 (Otto's translation). 


July 1208, in the Tail of the Scorpion, said to have been 
in aspect like Saturn. 

1280, in Ophiuchus. 

1672, Tycho's star in Cassiopeia. 

1604, Eepler*s star in Ophiuchus, 

1670, in Vulpecula. 
' 1848, in Ophiuchus. 
" 1860, in Scorpio. 

1866, in Corona Borealis. 

1876, in Cygnus. 

1885, in Andromeda. Making eighteen in all, besides four 
or five questionable instances mentioned in Chinese annals. 

The most noteworthy feature of this list is the curiously 
partial distribution of the objects enumerated in it. Nearly 
all of them lie in the thoroughfaxe of the Milky Way, and 
half are clustered together in the section of it marked by the 
stars of the Scorpion and the Serpent-tamer. In time also 
the grouping of the apparitions is strikingly unequal. The 
occurrence of three within the seven years 886 to 898 a.d. 
was succeeded by a blank of four and a half centuries. 
Kepler's came pretty close upon Tycho's star ; and no less than 
five * Novae* blazed out during thirty-seven recent years, while 
the latest previous record of the kind was of Anthelm's star 
in 1670. 

The brightest sidereal object known to us by authentic 
description was the * stranger-star ' in Cassiopeia, observed by 
Tycho Brahe. He first saw it November 11, 1572, but it had 
already been noticed by Lindauer at Winterthiir, November 7, 
and Maujrolycus entered upon its systematic study at Messina 
November 8. From equality with Jupiter it rose in a few days 
to be the rival of Venus, showing to keen eyes at midday, and at 
night through clouds thick enough to obscure every other star. 
After about three weeks, however, it began to fade, and in 
March 1574 disappeared finally. Its colour was at first 
dazzlingly white, then for a while ruddy, and from May 1578 
onward, pale with a livid cast. Bapid scintillation distin- 

" Wolf, Qeschichte der Astronomie, p. 414 ; Kaiser, De Sterrenhemel, Part i. 
p. 582. 


guished it throughout.^ There is no reason to suppose its 
outburst other than solitary. The appearances in the years 
946 and 1264 connected with it by a Bohemian astrologer 
named Cyprian Leowitz,' were almost certainly apocryphal.' 

The * new ' star (designated ' B Gassiopeiffi ') can still be 
perceived smouldering in the spot where it once blazed. 
Tycho's measurements, reduced and discussed by Argelander, 
located it within one minute of arc of a reddish, eleventh- 
magnitude star, first noticed by d'Arrest in 1865, the character 
of which, as disclosed by the observations of Hind and Plummer 
in 1870-4,* fully warrants the inference of its identity with 
the famous ' temporary.' Not only is it variable to the extent 
of nearly a magnitude, but it frequently seems nebulous, with 
occasional lurid flashes of momentarily increased brightness. 
Its non-appearance in a photograph taken by Mr. Boberts 
January 12, 1890, showing above 400 stars where d' Arrest 
charted 212,^ may be due to the actinic feebleness of its light. 

The star of 1604 ran a parallel course to that of 151^ 
Discovered by John Brunowski October 10, it quickly over- 
topped Jupiter, but by the end of March 1605 had sunk to the 
third magnitude, and a year later vanished. Kepler describes 
it as sparkling like a diamond with prismatic tints,' ^ but says 
nothing of progressive changes of colour. • Nova Serpentarii ' 
has left behind no clearly identifiable representative. 

The next * new star ' was discovered near fi Cygni on June 
20, 1670, by Anthelmus, a Carthusian monk at Dijon. It was 
then of the third magnitude, but its decline, unlike that of 
others of its class, was interrupted by two reappearances 
separated by intervals of invisibility. Between March and 
May 1671, it rose from the fourth to the third rank, then died 
out, only flickering up to the sixth magnitude in March 1672.^ 
Almost exactly in its assigned position, Mr. Hind picked up, 
April 24, 1862, a star between the tenth and eleventh magni- 

> Tycho, De Novd Sielld anni 1672, p. 302. * Judicium de Novd SteUd, 

* Lynn, Observatory, vol. Ti. pp. 126, 151 ; Sadler, English Mechanic^ yoI. 
xzz. p. 402 ; Tycho Brahe, Progymnasmataj p. 831. 

* Monthly Notices^ vol. zzxiv.p. 168 ; Gore's Catalogue of known Variables, 
p. 164. • Monthly Notices, vol. 1. p. 359. 

* Kepleri Opera, t. ii. p. 620. * J. Cassini, EUments d'Astronomie, p. 69. 


tude, which, when reobserved in 1861, had lost more than 
half its light, and gave the blurred image characteristic of 
many superannuated ' Noysb.* ^ The triple maximum of 
Anthelm's star assimilates it to Janson's variable P Gygni,'^ 
which has itself often been classed as a Nova. 

Undoubtedly such was an object detected by Mr. Hind in 
Ophiuchus, April 28, 1848, when it was of 6*7 magnitude, 
and intensely reddish yellow.' Four days later it had 
mounted above the fifth magnitude, from which eminence it 
slowly descended, making no lasting halt until, in 1874-5, it 
had got down to the thirteenth magnitude.^ 

With the spectroscopic study of temporary stars, a fresh 
chapter in our knowledge of them opened. Through the 
magic of the prism, more was ascertained as to their essential 
nature in five minutes than could have been learned in as 
many centuries with the telescope alone. On May 12, 1866, 
Mr. John Birmingham, of Millbrook, near Tuam in Ireland, 
^as amazed to perceive an unfamiliar star of the second 
magnitude shining in the constellation of the Northern Crown. 
On May 16, the application of Dr. Huggins's spectroscope 
showed the object to be wrapt in a mantle of blazing hydrogen. 
Five bright hnes (three of them due to hydrogen) stood out 
from a range of continuous light, broken up into zones by 
flutings of strong absorption.'^ The incandescence of the star 
was hence largely atmospheric, and for the rest, from the rapid 
rate at which it fell away, could have been only * skin-deep.' 
That the compound nature of its spectrum testified truly to an 
immense diffusion of vaporous material in its neighbourhood 
was certified by Dr. Huggins's visual observation of a singular 
glow round the star on May 16 and 17. Although its light 
decreased by a daily half magnitude, and its colour changed 
from white to orange, no alteration took place in the character 
of the spectrum. The bright rays, however, faded somewhat 
less promptly than the continuous light. 

* Monthly Notices^ vol. zxi. p. 231 ; Nature, vol. zzxii. p. 855. The star is 
No. 1814 in the Greenwich Catalogue for 1872. 

-- ' Bee anU, p. 70. ' Astr. Nach. Kos. 6S6, 638, 672. 

* Monthly Notices, yoI. xzi. p. 232. 

* Proceedings Royal Society, vol. xv. p. 146. 

H 2 


The visibility of the object to the naked eye lasted only 
eight days, and already, in the beginning of June, it had sunk 
io the ninth magnitude. Its slow subsequent decline was 
interrupted by fluctuations, thought by Schmidt to be periodical 
in about ninety-four days.* When observed by Vogel, March 
28, 1878, it was of the tenth magnitude, and gave an ordinary 
ftellar spectrum.' Virtually, it had resumed the conditions 
of its existence when Schonfeld entered it as of 9*5 magnitude in 
the ' Bonn Durchmusterung.' Its leap upward to the second 
magnitude, involving a thomand-fold gain of light, was ac- 
complished with extraordinary suddenness. Two hours and 
a half previously to Birmingham's discovery, Schmidt 
surveyed at Athens the constellation in which the blaze was 
about to occur, and noticed nothing imusual. He was certain 
that the star could not then have been as bri^t as the fifth 

The name of ' T GoronaB * was bestowed upon it in con- 
formity with Argelander*s system of nomenclature, by which 
the variables in each constellation are designated, in the order 
of their discovery, by the Roman capital letters from R onward. 
X)nly stars otherwise anonymous, however, are included in the 
distinctive series thus created, so that many variables are still 
entitled in the ordinary way by Greek letters. 

The stellar apparition that ensued after ten years was, in 
some of its features, the most remarkable of all. Dr. Schmidt 
noticed at Athens, November 24, 1876, a star of the third 
magnitude near p Cygni, in a spot till then vacant, so far 
as recorded observations went. The weather having been 
cloudy during the previous four days, there was no possibility 
of tracing the steps of its ascent, but it ran down very rapidly, 
and ceased to be visible to the naked eye on December 15. 
Its changes of colour pursued an inverse order to those of 
its predecessor. From golden yellow it turned white, and 
eventually bluish. 

The earliest spectroscopic examination of *Nova Cygni* 
was made by M. Cornu, at Paris, December 2 and 4.^ Just 

• • Astr, Nach. No. 2118. « MonatsberichU, Berlin, 1878, p. 304. 

" Comptes Bendtis, t. Ixzxiii. p. 1172. 



the same range of bright lines was measured by him which 
would start into view in the solar spectrum upon a considerable 



Deo. 8, 1876. Deo. 14. Jan. 1, 1877. Feb. S. llarohS. 

Fio. 10.— Changes in the Speotrom of Nova Cygni (Vogel). 

augmentation of incandescence in the sun's gaseous sur- 
roundings. Besides three, if not four hydrogen lines, there 
were the green coronal ray (wave length 5816), the yellow 


helium ray (wave length 6875), and the magnesinm group b. 
With these were very curiously associated the fundamental 
nebula-line at 5005, and possibly a line at 4476, long after- 
wards detected by Dr. Copeland in the spectrum of the Orion 
nebula* The changes affecting the light emissions of the new 
star between December and March, shown in fig. 10, from 
Yogel's drawings, indicate the remarkable transformations 
undergone by the fading object.^ At the height of its outburst 
it might be called of the type of fi Lyr» and 7 CassiopeiaB. 
The C of hydrogen was vivid ; the continuous spectrum strong. 
Gradually G yielded its supremacy to F ; only a faded remnant 
of the general prismatic light survived in the yellow and blue ; 
D3 (helium) vanished ; ' and the two leading rays of 7 Argus 
made an unexpected appearance. First determined by Cope- 
land, January 2,' they became more prominent as the Nova 
declined, the likeness to the Wolf-Bayet group being enhanced 
by the presence of the intense absorption-band in the blue,^ 
common to them all. 

Meanwhile the nebula-line had been steadily creeping to 
the front ; and when observations, suspended in March owing 
to the encroachments of daylight, were resumed at Dunecht 
by Dr. Copeland, September 2, 1877, it stood alone.' All the 
remaining light of the object (which had by that time sunk to 
10*6 magnitude) was concentrated in that one green ray, and 
a stellar or planetary nebula was, to all appearance, substituted 
for a star. This phase, however, proved scarcely less transient 
than the rest. So far as could be made out with the Harvard 
fifteen-inch refractor in 1880 (when the Nova gave only about 
one-seventh as much light as in 1877), the spectrum was that 
of an ordinary star.^ A confirmatory negative observation was 
made at Dunecht, February 1, 1881. Nothing could then be 
seen spectroscopically of the nondescript object ; the effect of 
the prism was to extinguish its faint rays. This proved that 
it dispersed them — in other words, that they were of various 

■ See Lockyer, Proe. R. Society, vol. xliii. p. 189. 

* It was last seen by Copeland, Jan. 9, CopemicuSt vol. ii. p. 111. 

* Ibid. pp. 102, 112. 

* Monatgberiehie, Berlin, 1877, p. 255 (Vogel). 

* Copmnieui, vol. ii. p. 106. • Atmual Report, 1879-60, p. 7. 


refrangibilities. For had they been monochromatic, the image 
that they gave would have remained intact. The latest 
observations of Nova Gygni were made by Mr. J. 6. Lohse in 
1885-6, witti the 16^-inch refractor of Mr. Wigglesworth's 
observatory near Scarborough. They showed it as of fourteenth 
or fifteenth magnitude, bluish in colour and nebulous in aspect. 
No spectrum of any kind could be obtained from it.^ 

The steps by which the famous 'temporary' of 1876 
descended toward approximate extinction are the same by 
which, in Mr. Lockyer's scheme of development, cosmical 
masses ascend from a nebulous towards a truly sun-Uke con- 
dition. It is indeed unproved, and perhaps unprovable, that 
these remarkable migrations are repeated, in an inverse order, 
as a normal result of condensation ; but it is scarcely doubtful 
that their relations of temperature have at any rate been 
rightly assigned. The Nova was certainly at its hottest when 
the emissions of helium were conspicuous in its spectrum; 
the * Swan-star ' lines emerged upon a considerable advance 
in cooling, a further stage of which was marked by the dis- 
appearance of hydrogen, and the solitary survival of the 
nebula-line. This, be it remarked, was visible from the first, 
BO it seems to stand all temperatures. Nor was its display 
in the midst of a cosmic conflagration wholly without prece- 
dent. The spectrum of T Coronas, there is reason to believe, 
included it,' as well as the ill-defined blue band distinctive 
of stars on the spectral pattern of 7 Argus. 

The two outbreaks were then of essentially the same 
character, notwithstanding some variety in the phenomena 
ensuing upon them. The difference may perhaps be explained 
by the precipitous nature of the descent of T Coronae. The 
orderly sequence of changes undergone by Nova Cygni did 
not get time to develop in the earlier object. All qualities of 
light radiated by it faded with nearly equal rapidity. 

The N0V8B of 1866 and 1876 were not more clearly set 
apart together by the identical peculiarities of their light, 

' Monthly Notices, yoI. xIyu. p. 494. 

* See Vogel'8 reductions {Monatsberickte, Berlin, 1877, p. 242) of Stone's 
and Carpenter's measures {Monthly Notices, yoI. xzvi. p. I 


than the pair we are now abont to describe by the extraordinary 
circumstances of their situation. On May 18, 1860, a nebula 
in Scorpio, numbered 80 on Messier's list (6098 in Dreyer's 
New General Catalogue), was observed by Dr. Auwers at 
Berlin.' It presented its usual appearance of a somewhat 
hazy ball of light, brightening gradually inward, and resolvable 
with difficulty into separate stellar points, together constituting 
a closely-packed, and most likely excessively remote globular 
cluster. Three nights later he looked again, and saw that 
these minnows had a triton in their midst. A seventh magni- 
tude star shone close to the centre of the stellar group* The 
existence of the new-comer lasted visibly just three weeks. 
Before May 25 a decline set in ; it had made considerable 
progress when, on May 28, Mr. Pogson (uninformed of 
Auwers's discovery) was * startled* by the apparent svbsti- 
tutian of a star for the nebula,' the dim luminosity of which 
seemed actually obliterated by the keen stellar radiance 
emanating from within it. It recovered, however, very speedily 
from this merely optical effacement. On June 10 its normij 
aspect was almost restored, and has never sinee been dis- 

But after the lapse of a quarter of a century, the signifi- 
cance of this event was accentuated by the occurrence of a 
similar one elsewhere. This time the great nebula in the 
girdle of Andromeda was the scene of the outbreak. The 
unlooked-for addition to it of a ^ star-like nucleus ' was an- 
nounced by Dr. Hartwig at Dorpat, August 81, 1885 ; but it 
turned out that the change had already been perceived by 
Mr. Isaac W. Ward of Belfast, August 19, and two nights 
earlier at Bouen by M. Ludovic Gully, who, however, set it 
down as an effect of bad definition.' Concordant observations 
by Tempel at Florence, Max Wolf at Heidelberg, and Engel- 
mann at Leipzig showed decisively that the strange object 
made no show down to 10 p.m. on August 16 ; ^ and a photo- 
graph taken by Mr. Common in August 1884 gave positive 
assurance that its place had then no stellar cnscupant as bright 

1 Astr. Nach, No. 1267. ' Monthly Notices, vol. xxi. p. 32. 

" Ciel et Terre, Oct. 1, 18S5. ^ Astr, Nach. Nos. 2682. 2683, 2691. 


as the fifteenth magnitude.* What were virtually the first 
rays of the Nova reached the earth August 16, 1885. 

Between that date and August 81, it mounted from the 
ninth to the seventh magnitude ; then without delay entered 
upon nearly as swift a downward course, checked, however, 
by one decided pause. Even the largest telescopes failed to 
keep it in view after March 1886. The full yellow colour, by 
which the star at first contrasted efiEectively with the greenish 
nebular background it was projected upon, faded with its 
light. No haze or glow blurred its image, which remained 
sharply stellar with a power of 1100 on the great Princeton 
refractor, when the adjacent nucleus of the nebula melted 
into a confused luminous blot.^ Attempts, incomplete from 
the nature of the case, made by Dr. Franz at Eonigsberg; 
and by Professor Hall at Washington, to determine the parallax 
of Nova AndromedsB, gave only negative results.' So far ap 
they were significant at all, they indicated its immeasurable 
remoteness from the earth ; nor should it be overlooked that 
Sir Robert Ball's similar experiment upon Nova Cygni had 
suggested a similar conclusion.^ 

The spectrum of Nova AndromedsB was of a dubious 
character. It bore witness to a completely different order of 
incandescence from that of the ' blaze stars ' in the Northern 
€rown and the Swan. The bright rays which it probably 
included were inconspicuous. Dr. Huggins, on September 9, 
was nearly sure of the presence of several in the green and 
yellow ; * and Dr. Copeland succeeded with difficulty, on Sep- 
tember 80, in getting rough measures of three accessions of 
light,® one of them near the place both of the chief carbon- 
fluting (at 6164), and of a * maximum ' measured by Mr. Taylor, 
long after the disappearance of the star, in the spectrum of 
the Andromeda nebula. Besemblance in quality of light was 
indeed one of many arguments proving the physical relation- 
ship of the two objects. This was, however, superfluously 

^ Nature^ vol. zxzii. p. 522. ' Young, Sidereal Messenger, vol. iv. p. 282. 
■ Astr. Nach. 2816. * Duneink Observations, Part V. p. 24. 

* Beport Brit. Association, 1885, p. 936. 

* Monthly Notices, vol. zlvii. p. 54. 


certain. It is barely conceivable that one stellar conflagration 
should by chance be projected aknost accnrately upon the core 
of a nebula in reality quite disconnected from it; but that 
two such highly improbable events should occur within 
twenty-five years of each other may fairly be called im- 
possible. The Nov» of 1860 and 1885 were then each 
situated within the substance of the nebulaB they temporarily 

This collocation obviously falls into line with the galactic 
affinities of other temporary stars. All of them, except the 
apparitions of 1012 in Aries and of 1866 in Corona, were 
Milky Way objects. Now the Milky Way is a plane of con- 
densation for all small stars, but more especially, and in a 
marked degree, for stars as well as nebul» of a gaseous nature. 
Temporary stars are closely cognate with these, not merely 
through the brief gaseous incandescence bringing them to 
our notice, but through the symptoms of nebulosity which 
survive it. They are not ordinary, full-grown suns over- 
whelmed by some sudden catastrophe. The catastrophe, if 
determined by some external event, is prepared and rendered 
possible by their own peculiarities of constitution. It is true 
that a sun like ours might be led, in the course of its onward 
sweeping through space, to traverse one of the vast diffused 
nebulosities with which the heavens abound, when a conver- 
sion of a large amount of its motion into heat might, with 
formidable results, be expected to ensue. But the occurrences 
hitherto actually observed are of a different kind ; and although 
a plausible way of accounting for them is at hand, it is one 
not to be admitted without great reserve. 

Collisions in some form seem, at first sight, the only ex* 
planatory resource available. But we must distinguish. An 
encounter between two condensed masses (apart from its 
extreme improbability) would produce quite other, and more 
permanent, effects than those connected with the outbreak of 
new stars. Such transient flashes cannot be due to bodily 
collisions; only atmospheric collisions, minutely exemplified 
in the gleam of every shooting star that gets entangled in our 
upper air, can be concerned in them. Substantially this in- 


ference was arrived at by Mr. Lockyer in 1877, from a study 
of the appearances presented by Nova Gygni. 

* We seem/ he then remarked, ' driven from the idea that 
these phenomena are produced by the incandescence of large 
masses of matter, because, if they were so produced, the run- 
ning down of brilliancy would be exceedingly slow. Let us 
consider the case then on the supposition of small masses of 
matter. Where are we to find them ? The answer is easy : — 
in those small meteoric masses which an ever-increasing 
mass of evidence tends to show occupy all the realms of 
space.' ^ 

In his recent cosmical scheme, accordingly, the far-away 
cataclysm represented to our senses by the appearance of a 
* new star,' takes shape as a * collision between two meteor- 
swarms.' If we call the smaller of the two a comet, the 
larger a nebulous star, we shall get rid of much that is hypo- 
thetical, and may succeed in realising the situation more dis- 
tinctly. It is evident that nebulous stars must be more 
subject than others to encounters of the kind. Our sun, if dis- 
tended so as to fill the orbit of Mercury, would engulf in- 
numerable comets that now slip clear round it at perihelion. 
In past times, it probably has absorbed hundreds, nay 
thousands of these bodies. Enormous comets, moving with 
high velocities towards bodies (consequently) of great attrac- 
tive power, should indeed be called into action to produce the 
conflagrations of * new stars ; ' but there is no reason known 
to us why these conditions should not be fulfilled. We must 
recall, however, that such conflagrations only exaggerate the 
changes of other variable stars, and should accordingly be 
referred to an intensification of their cause. And since that 
cause can scarcely (as we shall see in the next chapter) be 
found in actual collisions, the doubt arises whether we are 
entitled to assume their occurrence in the case of temporary 
outbursts. If not, then we should substitute for them grazing 
encounters with nebulous masses revolving in hyperbolic orbits, 
and overthrowing by their proximity to the attractive body a 
thermal equilibrium already eminently unstable. 

* Naturct yoL zvi. p. 418. 




About two hundred and fifty stars have been formally regis- 
tered as variable, and many more are open to the like sus- 
picion. Gore's * Revised Catalogue'^ includes 243 entries, 
besides 89 provisional additions ; Chandler's neaiiy contem- 
poraneous list^ enumerates 225 objects. Of these 160 are 
reckoned as ' periodical/ the rest as * irregular ' or * tempo- 
rary.' Periodical stars are further divided into those with 
* long,' and those with ' short ' periods. Nor is the distinc- 
tion by any means arbitrary. The stars seem to separate of 
themselves into two principal groups, undergoing fluctuations 
in cycles of respectively less than fifty, and between two and 
four hundred days. The paucity of stars with periods of in- 
termediate lengths is shown graphically in fig. 11, where the 
height of the curve represents the numbers of stars subject 
to changes proportionate in duration to the horizontal dis« 
tance from left to right. 

Variations requiring several months for their completion 
differ both in degree and kind from those run through in a 
few days. They are of much greater amplitude, ranging over 
five to eight instead of, at the most, two magnitudes ; they 
are accomplished with less punctuality; and they are fre- 
quently attended by symptoms of atmospheric ignition en- 
tirely foreign to quicker vicissitudes. Most important of all, 
they affect bodies of pecuUar constitution. Nearly all long- 
period variables are red stars with banded spectra ; those of 
short period are white or yellowish in colour, and display 
Sirian or solar spectra. Quality of light is thus the pre- 

» Proc. R. Irish Acad, Vol. i. ser. iii. p. 97. * Astr, Journal, Nos, 179-180. 


dominant factor in determining the law of stellar light- 
changes, and is itself dependent, as we have seen, upon at- 
mospheric conditions. Hence we reach the generalisation 
that sUght absorption accompanies short, strong absorption 
long periods, and that extensive gaseous surroundings not 
only favour variability in general, but almost absolutely pre* 
scribe its type. 

Periods of between one and two hundred days may be 
called ^ long ' ; but as fig. 11 shows, they do not commonly 


Uudet 30« 60 90 120 160 190 SIO 240 270 300 380 360 390 420 450 480 

Fia. 11.— Distribution of 171 Periods of Variable Stars from 
Gore's Revised Catalogue (1888). 

occur. Such fluctuations as now engage our attention usually 
demand more than two hundred days for their accomplish- 
ment, and are seldom prolonged beyond 410. Mr. Chandler 
considers 820 days as the average duration of change for 
long-period variables ; ^ the prevalence, however, among them 
of periods of about one year is remarkable, and cannot be ac- 
counted for by mere accidents of observation. The first and 

> Astr, Journal, No. 193 


best known epecimen of the class anticipates by about a 
month the role of annual recurrence. 

When Bayer, in 1603, affixed in his charts the Greek 
letter o to a small star in the neck of the Whale, he had no 
suspicion of its identity with a supposed ' Nova ' which had 
disappeared seven years previously, after blazing up to the 
second magnitude. Its discoverer, on August 13, 1596, was 
David Fabricius, of Osteel, in East Friesland ; but though he 
saw the object again February 15, 1609, he left it to John 
Phocylides Holwarda, Professor of Philosophy at Franeker in 
Holland, to ascertain its true character in 1639 ; and the re- 
petition of the phases once in 333 days was established in 1667 
by Boulliau.^ The name * Mira ' bestowed by Hevelius upon 
the changing star in Getus, commemorates the amazement 
excited by the detection of steUar periodicity. 

The phenomena it presents would seem incredible were 
i they less well established. Once in eleven months the star 
; mounts up in about 110 days from below the ninth often to 
the second magnitude or even higher ; then after a pause of 
two or three weeks, drops again to its former low level in 
twice the time, on an average, that it took to rise from it. 
The brightest maximum on record was observed by Sir 
William Herschel, November 6, 1779, when Mira was little in- 
ferior to Aldebaran ; ^ the faintest minimum, that of 1783, is 
said to have carried it below the tenth magnitude. An extent 
of eight magnitudes may then be assigned to the oscillations 
of this extraordinary object, which accordingly emits, at cer- 
tain times, fully fifteen hundred times as much light as at 
others. That each maximum is a genuine conflagration has 
been proved by spectroscopic observation ; and the conflagra- 
tions recur yearly, with approximate regularity, and, after 
three centuries of notified activity, give no signs of relaxation ! 
The height of the maxima, however, varies greatly. The 
two adjacent ones of 1885 and 1886 (represented in fig. 12) 
showed a nearly fivefold difference of intensity; but Heis's 
remark that high and low maxima tend to alternate, has not 
in the long run proved consonant with facts. There is no 

> Monitum ad Astrotwrnos, p. 7. ' PhU, Trans, vol. Ixx. p. 338. 


role by which the brilliancy of impending phases can be pre- 
dicted. That of November 1868, in which the star just failed 
to reach the fifth magnitude, was, it is true, preceded by a 
high maximum, but several average or low maxima followed 
it. An that can be said is that exceptionally bright apparitions 
are isolated ; they do not come in sets, but one by one, at 
considerable intervals. 

The minimum brightness of Mira is about 9*5 magnitude, 
and tolerably uniform ; though Schonfeld stated in 1875 that 
he had never seen the variable inferior to its ninth magnitude 
companion.^ Its redness during low phases, however, em- 
barrasses estimation. Their nature, too, is to a great extent 

During the four months of each period that the star is ' out 
of sight,* it is apt to slip * out of mind ' as weU. Most pro- 


Montha 1 S 8 4 6 6 7 8 9 10 11 IS IS 14 1ft 16 17 18 19 SO 21 99 

Fio. 12.— Maxima of Mira in Febrnaiy 1885 and January 1886. 

bably, perceptible change is suspended for at least a couple of 
weeks before and after minimum ; but details as to this in- 
terval of inertness are wanting. Yet it would be of especial 
interest to ascertain its length, and to note the symptoms in- 
dicating its termination. How does recovery set in? is a 
question that has yet to be answered. Is it with a sudden 
start, or by a gradual revival ? Is it accompanied by changes 
of colour, or of spectrum ? Does telescopic definition remain 
the same after as before the critical epoch ? All these points 
have a theoretical importance likely to grow with time. 

The periodicity of Mira obeys a highly complex law. De- 
viations to the extent of a fortnight from the mean period of 
881 days are common, and the maximum of September 29» 
■ Jahreshericht Mannheimer VeremfUr Naturkunde, Bd. si. p. i9^ 


1840, was a fall month late.^ Its perturbations are, indeed, 
probably themselves periodical, but so many exist which have 
so far not been formulated, that prediction is often at fault. 
Argelander detected the influence of a wave of disturbance 
with an amplitude of twenty-five days, and embracing eighty- 
eight periods ; ^ Schwab's observations indicated subordinate 
oscillations in six and a half days ; ' and there are half-efiaced 
traces of several besides/ The shape of the light-curve, too, 
varies notably. Its peaks are sometimes much blunter than^ 
at others ; and the star, which usually retains its full lustre 
during a fortnight, has been known to remain twice that time 
stationary. Still more singularly, the otherwise invariable 
rule of an increase more rapid than the ensuing decrease was 
reversed in 1840. Sixty-two days were occupied in ascending 
from the sixth to the third magnitudes, forty-nine only in 
sinking back to the same level. The anomaly, due to the 
retarded maximum of that year, recalls the abnormal course of 
the sunspot cycle which culminated at the close of 1883. 

The spectrum of Mira is a splendid example of 8ecchi*s 
third type. Eleven bands of profound shadow, sharp towards 
the violet, gently gradated towards the red, throw out into 
strong relief the intervening brilliant zones ; while dark lines 
of metallic absorption, and vivid hydrogen-rays, vary the 
effect, and add to the intricacy of the characters to be de- 
ciphered. Only the more refrangible members of the hy- 
drogen series appear to be brightened in this star ; no trace 
of G or F betrays itself to the most attentive scrutiny. Yet 
they can hardly be in reality absent. We can only suppose 
that these slower vibrations are partially absorbed by over- 
lying strata of cooler hydrogen, which, by reason of their lower 
temperature, are inactive as regards the quicker vibrations. 
The detailed study of the bright lines that do appear can 
best be carried on by photographic means. It may be as- 
sumed that they become extinct with the approach of each 
minimum, and are re-kindled during the ascent towards the 

> Argelander, Astr. NacK No. 416. 

< Bonner Beobaehtungtn, Bd. vii. p. 832. ' Astr, Nock, No. 2731. 

* Aigelander in Humboldt's Cosmoa, yol. iii. p. 234. 


Bnsamg maximum ; but positive evidence to this effect is still 
a desideratum. Absorption certainly increases with the fading 
of the star. The spectral bands, though very intense at 
maximum, deepen and widen remarkably after it is passed.^ 
The diminution of brightness, however, is not entirely due 
to this cause. Light intrinsically fails, besides being addition- 
ally intercepted. 

A reddish-orange star of the sixth magnitude was dis- 
covered by Mr. J. E. Gore, of Ballysodare, in Ireland, on 
December 13, 1885. Its situation, just where the Milky Way 
streams across the club of Orion, might have seemed con- 
firmatory of the temporary character imputed to it mainly on 
the ground of the improbabiUty of so bright an object having 
so long escaped notice. But the decline which carried it 
below the twelfth magnitude in July 1886 was succeeded by 
a renewal of light, and a second maximum occurred within 
a day of the anniversary of the first. The star has since 
conformed pretty regularly to a period of about a year, not 
always, however, filling its full measure of change. This, 
indeed, is the case with all variables of the Mira-type, to which 
* U Orionis ' (a name substituted for its original designation 
of *Nova Orionis') unmistakably belongs. Its spectrum 
might be called a replica of that of Mira, and both include the 
same photographic bright lines. The helium ray in addition 
was probably seen by M. von Konkoly in the Orion variable. 

The history of the two stars has also points in common. 
Each, taken at first for a ' Nova,' was only on further acquaint- 
ance recognised as periodical* Each, too, unaccountably 
escaped astronomical notice for centuries ; yet it is perhaps 
easier to accept the usual explanation of this difficulty by a 
series of coincidences between epochs of minima and epochs 
of observation, than to suppose each star to have newly 
entered upon its vicissitudes at the time of its discovery. 

The second long-period variable recognised was a star in 

the neck of the Swan, which Bayer, ignorant of its changing 

character, set down in his maps as of the fifth magnitude. 

It still retains the name he gave it of * ^ Cygni.' Missed by 

» Maander, Monthly Notices, toI. xlix. p. 303. 



Gottfried Kirch in July 1686/ it reappeared October 19, and 
subsequently disclosed to his vigilant watch fluctuations even 
wider than those of the * wonderful ' star in Cetus. It descends 
nearly to the thirteenth and rises to the fourth magnitude, 
sometimes indeed stopping short when barely visible to the 
naked eye, but more commonly remaining lucid for a couple 
of months. Nor is its course much better regulated as regards 
time. Errors up to forty days often attach to its phases, and 
the attempt to correct them by the introduction of cycUcal 
disturbances has proved only partially successful.^ The period, 
estimated at 402 days by Eirch, now averages 410. Olbers. 
noticed that it had been steadily lengthening down to 1818,^ 
and it is lengthening still ; the compensatory process antici- 
pated by him has not set in. As usual in such cases, the 
ascent to maximum is much more rapid than the descent from 
it, occupying at present about 186 days.* 

The spectrum of x ^Jg^^ ^^ colonnaded like that of Mira. 
But on May 19, 1889, when the star was near a marimum, 
Mr, Espin perceived evidence in it of direct radiation by 
hydrogen and heUum, confirmed by Mr. Taylor's subsequent 
observations at Ealing with Mr. Common's giant reflector. 
The absorption bands appeared at the same time to have lost 
their determinate character, and as the general light of the 
variable waned in the autumn, an immense number of bright 
lines, carefully observed by Dr. Becker at Dunecht, came 
visibly into view. 

In B Hydr88, too, the advent of maximum brightness is 
attended by a blaze of hydrogen.* The cycle of change by 
which this star oscillates from above the fourth to the tenth 
magnitude is rapidly shortening. In 1708 it extended to 500 
days, it is now comprised in 434.* The period (about 229 
days) of S UrssB Majoris, on the other hand, has lengthened 
notably since 1855 ; ^ while a sudden, very considerable abridg- 
ment of that of S AquilsB, towards the close of 1883, was 

> Miscellanea Berolinensia^ t. i. p. 206. 

* Banner Beob, Bd. vii. p. 336 ; Mannheimer Jahresbericht, Bd. zl. p. 110. 

* Schumacher'^ Jahrbuch, 1841, p. 93. * Gore's CcUalogue, p. 197. 
« Espin, Astr. Nach. No. 2889. • Chandler, ib. No. 2463. 

' J. Baxendell, jun. Journal Liverpool Astr. Soc. vol. iii. p. 62. 




followed by great irregularities, mth a tendency, on the 
whole, towards restoration of the stattis quo ante,^ Nor is it 
likely that any such disturbances will prove indefinitely pro- 

It is not easy to decide off-hand whether U Geminorum 
should rank as a periodical or as an irregular variable. 
Habitually of extreme faintness (below fourteenth magnitude), 
it gains light at uncertain intervals with marvellous rapidity, 
approaching, or even slightly overpassing, the ninth magnitude. 

24 Sfi 26 27 28 29 30 81 1 S 3 
Jan. 1884 Feb. 

1 2 8 4 6 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 
AprU 1885 

Fig. 13. — Two Types of Maximam of U Geminornm (Knott). 

Thus its brightness in February 1869 increased sixteenfold 
within twenty-four hours ; ^ and Mr. Knott estimated its mag~ 
nitude February 20, 1877, at 18"2 ; but twenty-six hours later 
at 9-8 I * Two types of maximum are, according to his obser- 
vations, exhibited by this star.^ In one the entire swing to 
and fro between the fourteenth and ninth magnitudes is 
completed in ten days, or even less ; in the other it occupies 
fifteen to twenty. A pause at the summit, in the latter case, 

" Baxendell, Ohseroatory, vols. ix. p. 122 ; x. p. 261. 

* Schdnfeld, Sirius, Bd. x. p. 62 ; Mannheimer Jahreshericht, Bd. xl. p. 85 
» Monthly Notices, vol. xxxvii. p. 279. 

* Journal Liverpool Astr. Society ^ vol. iii p. 11. 


with a tendency to a secondary minimnm, accounts for the 
difference. The light-curve takes more or less the form of a 
double peak with a saddle between (see fig. 13). 

Three of these extraordinary crises are, as a rule, under- 
gone yearly; but the star remained quiescent during 617 
days in 1860-1, while some of its maxima are separated by 
no more than 65 to 75 days. Usually undistinguished in 
colour, it was described by Hind on the night of its discovery 
by him, December 15, 1855, as shining * with a very blue, 
planetary light,' * and the peculiarity of its tint (resembling 
that of a coerulean-hued planetary nebula in Hydra) at the 
maximum of April 1881 suggested to Safarik the probability 
of a gaseous spectrum being derived firom it.* Neither 
Dr. Copeland's observations in 1880, however, nor Mr. Espin's 
in 1889, afforded more than the barest suspicion, of bright 
lines,^ accompanied in the latter case by persistent indications 
of a strong dark ' F ' ; so that hydrogen, at any rate, is not 
kindled into exceptional brilliancy as the steep ascent towards 
maximum is climbed. U Geminorum is one among a number of 
variable stars, presenting at times a hazy and ill-defined aspect ; 
but this may be a mere optical effect of the correction for colour 
of the telescopes employed to view them. 

We now come to that unique star, 17 Argus. Its actual 
appearance is insignificant. Invisible to the naked eye, its 
reddish colour and slightly superior brightness alone distin- 
guish it in the telescope from the crowd of small stars 
embroidering one of the finest of the southern nebulae, some- 
times named the ^Key-hole Nebula,' from the aperture of 
that shape with which it is centrally perforated. Close to an 
edge of the aperture, in the densest part of the nebula, 
97 Argus is placed. Nor can we suppose its position fortuitous. 
Although probably constituted for variability, its situation, 
plunged (as appears certain) in nebulous substance, combines, 
we are led to suppose, with its essential nature to produce 
the exceptional character of its changes. 

The first observation of 17 Argus was made by Halley at St. 

» Monthly Notices, vol. xvi. p. 56. « AsU'. Nach. No. 2391. 

» Observatory, vol. v. p. 110 ; Astr, Nach, No. 2919. 


Helena in 1677, when it was of the fourth magnitude ; the 
next by Pere Noel, a Jesuit missionary, in China, about ten 
years later.^ The second rank was assigned to it both by him 
and by Lacaille in 1751 ; yet the discrepancy with Halley's 
appraisement remained unnoticed. The higher estimate was 
besides confirmed by those of Fallows, Brisbane, and Johnson 
in 1822, 1826, and 1832 respectively, and only the traveller 
Burchell, familiar with the star as of the fourth magnitude 
in 1811-15, was surprised one night, at San Paolo in Brazil, 
to see it temporarily raised to a level with the finest brilliants 
of the sky. Another, and a still more vigorous outburst, was 
witnessed by Sir John Herschel at the Cape, December 16, 
1837. Without previous note of warning, the star all at 
once nearly tripled its light, and before the end of the year 
fully matched a Centauri. Since then it has been kept under 
strict surveillance as a notorious character, and not without 
reason. After a partial decline and several preliminary 

* flutterings,' it reached a final maximum in April 1848, when 
Sirius alone among the fixed stars slightly outshone it. This 
high position was moreover fairly well maintained for nine or 
ten years. Gilliss, at Santiago in 1850, found it very little 
inferior to Canopus in light, and in colour more deeply tinged 
with red than Mars.^ Still of the first magnitude in 1856,* 
it fell to the second in 1858, to the third in 1859, and ceased 
to be visible to the naked eye early in 1868.* 

For sixteen further years the slow ebb of light continued, 
and the magnitude of the once effulgent rj Argus, carefully 
determined by Mr. If inlay at the Cape, was in March 1886 
only 7'6.* This proved to be the lowest point touched. Slight 
symptoms of recovery became apparent in the two following 
years. * A glow and lightening of colour ' first attracted the 
attention of Mr. Thome at Cordoba, March 20, 1887.^ From 

* dull scarlet * the rays of the variable had vivified to ' bright 
orange.' And in May 1889 Mr. Tebbutt perceived and 

» Winneoke, Astr. Nach. No. 1224. 

* Abbott, Monthly Notices, vol. xxi. p. 230. • Mocsta, Astr. Nach. No. 1064. 

* Tebbutt, Monthly Notices, vol. xxxi. p. 210. 

* Monthly Notices, vol. xlvi. p. B40. 

* Astr. Nach. No. 292 


announced, from New South Wales, a decided increase of light,^ 
which has since been steadily maintained. Eta Argus stands 
now at about the seventh magnitude, and may before long 
resume its place among lucid stars. 

It is, nevertheless, far too soon to decide the question of 
periodicity. Quite possibly the history may be one that ' does 
not repeat itself.' Our continuous knowledge of it is embodied 
in the accompanying diagram (fig. 14), in which a single vast 
oscillation is indicated, occupying about a century for its 
completion, and diversified by secondary fluctuations of a very 
conspicuous character (innumerable minor ones are ignored 
in the figure). The data at present available, however, afford 
no grounds for concluding this oscillation to occur regularly. 

1810 18S0 1830 1840 1850 1860 1870 1880 1890 

Fw. 14.— Lightcurve of 17 Argda, 1810-1890. 

Attempts to assign a period to the variations of rj Argus have, 
so far, signally failed- Wolfs of forty-six,* and Loomis's of 
seventy years,* are both palpably too short ; an allowance of 
at least ninety is demanded by the tardy advent of the recent 
minimum. This would imply the occurrence of minima in 
1796 and 1706. But we have complete certainty from 
Halley's and F^re Noel's observations, that the star was on 
the rise towards the close of the seventeenth century ; and 
if its behaviour then resembled that observed in the nine- 

* Astr. Nach, No. 2849. * Monthly Notices, vol. xxiii. p. 208. 

» Ibid. vol. xzix. p. 298. 


teenth century, it must have continued to rise during at least 
twenty years after I6869 tlius reaching a maximum at a cal- 
culated epoch of minimum. A centennial or any longer period 
would encounter difficulties no less insurmountable, besides 
being in itself improbable, since no genuine stellar cycle has 
hitherto been found to embrace two complete years. Much 
more extended periods have sometimes been suggested, but 
never ascertained. The stars they were ascribed to, when the 
time came for a repetition of their presumed cyclical changes, 
showed a total want of conformity with what was expected of 
them. As examples may be mentioned 68 Gygni, to which 
Mr. Espin attributed a period of five, and B Gephei, thought 
by Mr. Pogson to obey one of seventy-three years. 

Thus a considerable future increase in the light of the 
great southern variable, but scarcely a revival of its past 
splendours, may be looked for. Two kinds of unsystematic 
stellar fluctuation can be distinguished. In one, a quickly 
compensated change, either upward or downward, takes place 
at uncertain intervals; in the other, the shifting from one 
to a higher or lower order of brightness is more or less 
permanent. A star may, under this regimen, not only by a 
sudden start double its light, but continue for years to shine 
with twofold intensity. We cannot even say that an eventual 
return to its former status is inevitable. Now in 17 Argus, 
features of both these methods seem to be combined. It has 
made quick springs, and held its ground, but it has also often 
kindled into evanescent brilliancy. Its stationary epochs 
have been followed by epochs of instability ; at some times it 
has shown a tendency to establish itself at halting*places, at 
others to slip along an inclined plane of change. In all this 
it differs materially from temporary stars, which leap up, as if 
by a single impulse, to their solitary maximum, after which they 
lose in a few months the whole of the light they had acquired. 

The range of variation of rj Argus is, indeed, without 
precedent except among new stars. It amounts to fully eight 
and a half magnitudes. When Sir Thomas Maclear observed 
this marvellous object in 1848, it was emitting 2,500 times 
as much light as when Mr. Finlay observed it in 1886 ! If 


asked to pronounce whether rising so high or sinking so low 
should more properly be regarded as an 'accident' of its 
strange career, we should be inclined to say the latter, since 
it figures as of second magnitude in most early observations.^ 
The visible spectrum of 17 Argus is not strongly characterised ; 
but photographs taken in Peru by Mr, J. E. Bailey, of Harvard 
College, show bright hydrogen lines, which were most likely 
seen with the great Melbourne reflector by Le Sueur in 1870.* 
Continued spectroscopic observations, as the star gains light, 
may prove of great value. 

Pickering's third class of variable stars may conveniently 
be made to include those subject to irregular fluctuations of 
every degree, and among them a variety described by Mr, 
Espin in 1887,^ of which the range is one and a half magni- 
tudes, and the mode of progression by quick and seemingly 
casual bounds. As examples, 19 Piscium and several other 
stars belonging to the fourth spectral type are mentioned. 
Closer attention has, however, been bestowed upon the slighter 
changes of some third-type brilliants. Sir William Herschel 
added in 1795 a Herculis to the list of seven variable stars 
then known.* But the period of two months which he 
assigned to its oscillations between 3*1 and 8*9 magnitudes 
has not been ratified. During some years they appear indeed 
almost to cease, then are hurriedly resumed, but with no settled 
order. The analogous variations of Betelgeux and /S Pegasi 
are equally unmethodical. 

The extraordinary character of a star long known as 
' Variabilis Coronas,' now called ' R CoronaB,' was discovered 
by Pigott in 1795 ; a near neighbour of the ' blaze star ' of 
1866, its changes are of the nature of extinctions rather than 
of outbursts. Ordinarily of the sixth magnitude, it occasion- 
ally drops out of sight with small telescopes, and after linger- 
ing near the thirteenth magnitude for many months, slowly 

* It is marked so in Bayer's charts (1703) probably on the aathority of 
Petrus Theodorus, who voyaged southward in 1595. Winnecke, Astr. Nach. 
No. 1224. 

« Travs. B. Society of Victoria, vol. x. pp. 11, 23. 

» Observatory, vol. x. p. 439. * Phil. Trans, vol. xxxvi. p. 452. 

O- ~ HE ^ 


regains its lost light. But its phases at times cease wholly, 
as during the seven years 1817-24/ at others are ill- 
marked. Thus, at the minimum observed by Sawyer, October 
13, 1885, the star was still of 7*4 magnitude.^ It shows no 
decided peculiarity either of colour or spectrum. 

R Cephei is a star which, since the beginning of the 
present century, has lost on an average |^ of its radiance, 
and at present in no way tends towards recovery. In the 
time of Hevelius it was of the fifth magnitude, and Groom- 
bridge's observation of it in 1807 showed it to be then still 
unchanged. By 1840, however, it had sunk to the tenth, 
and has never since risen above the eighth magnitude.^ Its 
identity with th^24 Cephei' of Hevelius, tracked out by 
Pogson in 1856,^ is universally admitted. Although its light 
was considered by Schonfeld to be tinged with red, it ap- 
peared bluish to Farley in 1838, and its analysis has disclosed 
no features of interest. 

Among genuinely red stars such shiftings of photometric 
standing are, perhaps, of ordinary occurrence. Of twenty-two 
such, kept in view by M. Bafarik, at Prague, from 1883 to 
1888, for the express purpose of testing their constancy, only 
nine remained without noticeable change, two were found 
periodically, six irregularly variable, and five either vanished 
or lost great part of their light. Earlier observations of 
several of these objects certified the progress of their decline 
during twenty to twenty-five years.® An example of a sudden 
acquisition of lustre is afforded by a small red star in the 
same field of view with y Cygni. Between December 1885 
and June 1886, Mr. Espin perceived it to have risen in rank 
by a whole magnitude,* that is, to be giving out two and a 
half times as much light as six months previously. And, so 
far as is known, the gain has been kept. 

We must now give some brief attention to the various 

* Argelander, Bonner Beob, Bd. vii. p. 874 ; Gibers, Berliner Jahrbuch, 
1841, p. 100. " Gore's Revised Catalogtu, p. 132. 

> Schdnfeld, Mannheimer Jahresbericht, £d. xl. p. 113 ; Gore's Catalogue, 
(1884), p. 200. 

* Monthly Notices, vol. xvii. p. 23. » Astr. Nach, No. 2874, 

* Journal Liv. Astr. Soc. vol. v. p. 2. 


explanations offered of the extraordinary phenomena described 
in this and the preceding chapters. Our task in this respect 
has been simplified by recent discoveries. The systematic 
occurrence of bright lines in the spectra of periodical stars 
near their maxima brings them, in the first place, into such 
close physical relationship with temporary stars as absolutely 
to prohibit the speculative separation of the two kinds of 
change they respectively exhibit. A theory stands self-con- 
denmed which deals with them on different principles. In 
the next place^ the association of variability with processes of 
luminous change in stellar atmospheres is now placed beyond 
doubt ; and this at once disposes of hypotheses of slag-forma- 
tion, nebulous eclipses, and axial rotation showing alternately 
bright and dark sides. Nor can we regard as admissible Dr. 
Brester's opinion that the sole cause of stellar fluctuations is 
to be found in intermittent chemical associations and dissocia- 
tions taking place at the atmospheric outskirts of cooling 
bodies.* For the increase of Ught at maidmum certainly 
ensues upon a real access of incandescence, and is not a 
mere appearance due to the dissipation of absorbing vapours. 
Apart altogether from apparitions of bright lines, M. Duner's 
observations discountenanced the supposition of a balanced 
inverse relation between the growth and decay of brilliancy 
and absorptive action in variable stars.^ They often proceed 
together, but not pari passu. 

Attempts have several times been made to explain the 
periodicity of stars through the influence of satellites revolving 
round them in highly eccentric orbits. Ellinkerfues suggested 
great atmospheric tides, raised at successive perihelion-passages,' 
as a means of bringing about periodic obscurations. Flass- 
mann's ^ view of tidal effects is wider, and perhaps embraces 
a partial truth. For, just as in the earth the unequal attrac- 
tions of sun and moon on its centre and surface sometimes 

' Essai d'vm Thiorie du Soleil et des Stoiles Variables^ Delft, 1889. 
' 8ur les J^toUea A Spectres de la Troisi^me Clause, p. 137. 

• QGttingische Nachrichten, 1866, p. 3 ; see also Dr. Wilsing's comments in 
Astr. Nach. No. 2960. 

* Die Verd7id€rlichc7i Sterne, Koln, 1888. 



provoke^ though they could not produce, earthquakes, so the 
tide-raising power of bodies making very close approaches to 
stars in a critical state of heat-equilibrium may serve as 
the occasions of luminous outbursts of a temporary or re- 
current nature. 

Mr. Lockyer*s meteoric hypothesis includes a novel ra- 
tionale of long-period variability ^ which deserves more serious 
attention than any previously suggested. Light-fluctuations 
are regarded in it as closely dependent upon stellar constitu- 
tion, and the bodies a£Eected by them as exceptionally diffuse — 


< / ,'i »-•'.•*'./ -. .' 

i • , • .• ^ ■ .. • 

• '*j:-t;>^;-'-'*c;.-v.s 

.•.* -^ . ■• ..■••.• '.•• : .••.,' « • • 

••.,.' '/-• , ,' .•.*•' 

. • • i 

^■^■.■<-r:A:r>^^:^; \ 

. ' . • 

•.••-/^>:-v'?:^Ji^g^.- V : 

* ••.*. * .• 

• ' t .' 


^^.' '•' 

.;, . ^';iyi^<rii^:^^Jr:\, ,\ 


*' ''/v^^4^-b^^ 

/ •' * 

••* - ' f'y^i^C'^' ''i* 





*LcV * • # ■ 

I ..•.r^/.V.^*- 

^^-t * 


V ^^^i0fi^ 



Fio. 16.— The Collision Theory of Variable Stars. 

postulates, both of them, fully warranted by observation. It 
has the further merit of treating temporary and periodical 
displays from the same general point of view, explains with- 
out constraint their quick development of light and more 
gradual decline, and fits in naturally enough with the gaseous 
blaze by which they are commonly attended. 

Its principle is briefly this. Variable stars are to be 
regarded as 'incipient double stars'; they are actually 
meteoric swarms with double nuclei, one of which moves 
swiftly round the other (see fig. 15), and at its nearest approach 

* 1?roc, 72. Societt/t vol. xliv. p. 80. 


dashes right through its outlying portions, with the result of 
innumerable collisions between individual meteorites, accom- 
panied by a vast increase in the combined light of the swarms. 
But the objection inevitably arises that this state of things 
could not long subsist. Even if set on foot, it should prove 
transient. By mechanical necessity, the satellite-swarm 
should speedily become extended into a ring, with, of course, 
complete effacement of variability. Thus each maximum of 
a star like Mira, if produced in the way supposed, would 
be feebler and more prolonged than its predecessor, until 
maxima and minima were brought to the same uniform 

The periodicity of variable stars is, besides, of far too 
disturbed a kind to be thus accounted for. Systemic stability 
would assuredly prove incompatible with the enormous irregu- 
larities it discloses. The abrupt acceleration or retardation, 
for example, by a month of the hypothetical attendant-swarm 
of Mira, would be impossible without such a total change in 
the elements of its circulation as would unmistakably break 
the continuity of its returns- 

But there are other objects far more recalcitrant than 
Mira to this mode of explanation. Take the outbursts of 
U Geminorum. They are not wholly capricious. There is a 
certain disorderly order about them by which they are mani- 
festly akin to the changes of more strictly periodical stars. 
We cannot then relegate them into a class apart, and invent 
a fresh hypothesis to suit them ; the collision theory, to be 
acceptable in the one case, must be capable of meeting the 
other. But we can scarcely conceive any combination of 
assumptions by which such an extension of its powers could 
be effected. 

Periodical cannot be sharply divided off from irregular 
variables. Every degree of disturbance, up to the total sub- 
version of laws of change, is met with among them. Many 
stars seem at times disposed to conform to a period which 
they later ignore. In others, method is indicated, though too 
vaguely to be defined ; while the majority oscillate, with wide 
allowance of amplitude, about a period itself often subject to 


periodical or secular change. It is evident that the imme- 
diate and unmodified interaction of revolving masses cannot 
explain breaches of regularity widening out to its total de- 

The time has not come to formulate a theory of stellar 
variability, but we may, at any rate, try to render our ideas 
on the subject coherent, and thus realise with some dis* 
tinctness the conditions under which alone any such theory 
could be regarded as adequate. 

As long ago as 1852, M. Budolf Wolf adverted to the 
analogous character of the curves representing sunspot fre- 
quency and stellar hght-change.^ They are not only of the 
same general form, but they are marked by precisely the same 
kind of irregularities. Both are steeper in ascent than in 

Scale of 

1867 1868 1889 1870 1871 187» 1878 1874 1875 1876 1877 
Fio. 16.— Curve of Sunspot Frequency, 1867-77 (Ellis). 

descent ; both rise into peaks of unequal heights at unequal 
distances apart. Mira, x Cygni, B HydrsB, and the rest, have, 
like the sun, retarded and accelerated, high and low, or abor- 
tive maxima. The representation in fig. 16, from the Green- 
wich observations, of the changes in sunspot frequency 
during the decennial period 1867-1877, is the very counter- 
part of the light-curve of a variable star. Especially character- 
istic is the break in the descending branch reflecting a partial 

» Mittheilungen Naturforsch, OeselUchaft, Bern, 1862, p. 261. 


recovery after maximam to which variables of all classes are 
prone. Moreover, the flow of the vicissitudes of stars and 
sun alike is broken and disturbed by the superposition upon 
the normal period of subordinate and superior cycles ranging 
from a few days, perhaps, to centuries. 

The presumption then of their similar origin is very 
strong; nor are we wholly without evidence of a physical 
nature to the same effect. The development of bright lines 
in the spectra of variable stars near their maxima is paral- 
leled in the sun by the increase of emissive intensity in the 
corona as sunspots increase. Atmospheric incandescence is 
thus in both cases heightened, although in immensely different 
degrees ; and confirmation is afforded to what was already 
certified by the congruous shapes of the two curves, namely, 
that the maximum of spots in the sun corresponds with 
the maximum of Ught in stars, and vice versd. It is the 
more necessary to bear this in mind because actual obscu- 
rations by spots have sometimes been alleged as a cause 
of stellar variability. The very opposite appears to be the 

The conclusion that solar and stellar disturbances are alike 
in kind, at least clears the ground for further investigation. 
For, besides obliging us to reject causes for the latter which 
are demonstrably unconcerned with the former, it renders 
sunspot studies directly available for solving the problem of 
stellar variability. Now what do we really know about the 
production of sunspots? Mr. Lockyer regards them, with 
much show of reason, as rents in the photosphere caused by 
downrushes upon it of cooled materials, these downrushes 
being themselves part and parcel of a great system of solar 
atmospheric circulation.^ They are, in fact, only local inten- 
sifications of the gentle rain falling always and everywhere 
over the surface of the sun, and indispensable doubtless to 
the maintenance of its luminosity. But of the alternate 
waxings and wanings of this mode of action, no complete 
explanation has so far been offered. Endeavours to connect 
them with meteoric influences have met with no success ; 
^ Chemistry of the Sun, chap, zzviu. 


the periodicity of the sun seems, however, to some extent 
dependent upon the situation of the planets. 

The likelihood, meanwhile, grows continually stronger, b,b 
the strange possibilities of close stellar combination are un* 
folded by new methods, that variable stars owe some of their 
peculiarities to complex modes of action upon them of satellite 
bodies. Mr. Lockyer may thus be right in supposing them of 
an essentially multiple nature, although the manner of regu* 
lation of their changes must be far less direct than that 
assumed by him. Tidal influences, largely appealed to by M. 
Plassmann, cannot be wholly inoperative, but had better be 
left to the future for detailed interpretation. The native ten- 
dency of red stars to instability of light is apparently connected 
with the great extent of their atmospheres, and we may further 
conjecture that the immediate cause of their luminous acces- 
sions is an enormous addition to the intensity of atmospheric 
reaction upon their photospheres. That is to say, prodigious 
falls of cooled matter take place, as in the sun, with the ap- 
proach of a spot-maximum, but on an immensely enlarged 
scale. Accompanying electrical phenomena are doubtless pro- 
portionately developed, and must exercise a powerful influence 
on the quaUty of luminous emission in such stars as Mira. 

It cannot be too strongly insisted upon, that while the re- 
currences of stellar light-change may be prescribed from with- 
out, its nature depends almost wholly upon atmospheric con- 
ditions. Extensively incandescent vaporous envelopes appear 
well-nigh inconsistent with real stability in shining. Where 
they are present, a trifling impulse may suffice to start an 
important disturbance. Thus a very large proportion of red 
stars are variable, and nearly all variables of long period are 
red. The length of the period, too, is very distinctly con- 
nected with the intensity of the colour. This was first 
noticed in 1878 by Dr. Schmidt, of Athens ; * it has been em- 
phatically confirmed by Mr. S. C. Chandler, who concludes, 
after an elaborate study of all the facts, that * the redness of 
variable stars is, in general, a function of the lengths of their 

» Astr. Nach. No. 1897. 


periods of light variation. The redder the tint, the longer 
the period.' ^ The redder the tint, also, the more profound 
(we are led to infer) the atmosphere, and the greater the 
distance, consequently, at which an attendant body should be 
situated in order to revolve free from it. This seems at least 
a possible reason for the correlation of duration of change with 
colour in variable stars of long period. 

■ Astr, Journal, Nos. 186, 193. 




Wb have seen, in the last chapter, that stars varying their 
light in periods of less than fifty days stand apart in several 
important respects from those undergoing slower changes. 
The distinction is accentaated by the tendency apparent in 
each class to group its members as far as possible from the 
frontier-line of separation from the other. Thus, long periods 
' for the most part exceed three hundred days, while a large 
majority of short periods fall below ten. Thirty-eight 
stars in all are reckoned as variable within fifty days ; of 
these thirty-two complete an oscillation in less than twenty, 
twenty-seven (including two with imperfectly ascertained 
periods) in less than ten days. A comparison of figures 17 
and 18 shows that, among short periods taken en masse, those 
of three to four days predominate ; those of five to eight days 
when Algol variables are excluded. 

Variables of short period are, as we have said, nearly all 
white or yellow stars. A very few are reddish ; and one — 
W Virginis — is suspected to possess a banded spectrum. 
B LyrsB, a star of about 4*5 magnitude, with a superb spectrum 
of the third type, is nominally variable in forty-six days, 
but its changes are so trifling and so imjierfectly rhythmical 
as to suggest that its proper place is with 13 Fegasi and 
a Herculis among stars affected by abortive periodicity. Those 
characterised by the description of variability we are now 
studying, seem to be bodies of more finished organisation than 
long-period variables, which (perhaps for that reason) they 
largely surpass in light-giving power. It has been remarked ^ 

» Espin, Observatory y vol. v. p. 79. 



that fluctuations are, on the whole, quick and Blight in pro- 
portion to the brightness of the objects they affect ; and quick 
fluctuations are executed with much greater precision than 
slow ones. In many stars the light ebbs and flows like 
clockwork as to time, and as to measure, with deyiations 
scarcely of the tenth of a magnitude from a settled standard. 
These remarkable changes progress gradually and continuously 
in Pickering's fourth class of variables ; in his fifth class they 

1 ^ 


2 *\ y 


o \/ 

09 S >r 

1 1 



ier . 1« 2 

3 4 

10 11 U 18 14 15 16 17 

19 30 

Fio. 17.— Distribution of all the periods of Variable Stars ander 
twenty days. 

only interrupt, although at perfectly regular intervals, the 
usually steadfast shining of certain stars. Of these two kinds, 
the former is conspicuously exemplified in yS LyraB — a star of 
which we have already made the acquaintance in connection 
with its gaseous spectrum — the latter in Algol. 

The variations of )8 Lyrae, detected by Goodricke in 1784, 
were first completely investigated by Argelander in 1844.* 
They are of a somewhat complex nature, including two equal 


10 11 IS 13 14 U 16 17 18 19 SO 

FiQ. 18.— Distribution of periods ander twenty days, excluding 
those of Algol variables. 

maxima separated by two unequal minima (see fig. 19). At 
its highest light the star is of 8*4, at its lowest, of 4*5 magni- 
tude. The intervening minimum is usually at 8*9 magnitude, 
but at times it tends to become ejBfaced, while at others it 
is scarcely less marked than the principal minimum. The 
intervals between all the four phases are approximately equal, 
but the curve of change is sharpest at the principal minimum, 

» De Stelld $ LijrcB Disquisitio. 


and it is then, accordingly, that the most exact observations 
are made. The entire cycle is traversed in twelve days, 
twenty-one hours, forty-seven minutes, and thirteen seconds, 
being some two hours and forty minutes more than the time 
occupied a century ago ; and the retardation continues to 
progress, though by no means uniformly. 

It is a singular fact that changes in quality are, in this 
star, combined with changes in quantity of light, yet in a per- 
fectly independent fashion. The bright lines in its spectrum 
vary seemingly on their own account. Instead of flashing 
out towards maximum, as in long-period variables, they, if 
anything, fade, as if overpowered by the general increase of 
lustre. Mr. Maunder suspected the hydrogen rays to be 
actually dark, October 19, 1888, when the star was at a 

0« 1 2 3 4 6 6 7 8 9 10 11 12 IS • 

Fio. 19.— Light-curve of /3 Lyrae (Argelander). 

maximum; while on the other hand, the only positive record 
of their showing by absorption was on September 5, 1882, 
the day after a secondary minimum. No rule, then, can be 
clearly made out, and an apparent anomaly has to be 
admitted. Beta Lyras is the only short period variable 
known to give a gaseous spectrum. 

Its double periodicity is reflected, though in a far less 
finished form, in B Sagittse, a star fluctuating comparatively 
slowly and unpunctually from about the tenth to S\ magni- 
tude. The period, normally of seventy days, shortens and 
lengthens alternately to the extent of some four days every 
ten years,^ and includes two unequal maxima and two u^equal 
minima (see fig. 20). The minima are subject to curious 
exchanges of intensity. They became equalised in 1873 with 

> Baxendell, Proc. Phih Soc. of Manchester, vol. xix. p. 120. 



the result of cutting the period completely in two ; * then the 
phase that had been subordinate grew to be the principal, 
while the principal declined to a subordinate rank ; and there 
are signs that the original state of things will, after a time, 
be restored. 

A pause in the decline, as if a second maximum were 

Mag. 8» 


Od ( 10 16 SO 35 30 . 35 40 46 50 55 60 6ft 

Fio. 20.— Light-curye of B SftgittsB (1884). 


contemplated but failed to be carried out, is a common cha- 
racteristic of short-period variables. It is well exemplified in 
S Cephei, which has continued, since its discovery by 
Goodricke in 1784, to oscillate with marvellous regularity 

0dl8846 0^1334667 

FiQ. 21.— Light-curve of 9 Cephei. Fio. 22.— Light-curve of v Aquilaa. 

between 8*7 and 4*9 magnitudes in a period of five days, 
eight hours, forty-seven minutes, and forty seconds. A swift 
ascent is accomplished in 1** 18-6**; then, sixteen hours after 
maximum, occurs the halt marked by the shoulder of the 
curve in fig. 21. The spectrum of S Cephei resembles that 

1 Schdnfeld, Vierteljahrsschrift Astr. Oes, Jahrgang xxi. p. 801 ; Baxendell, 
Proc. Phil. Soc. of Manchester^ vol. xxiv. p. 200. 


of the sun in being crossed by a great number of fine lines. 
A wide double star, its full yellow colour contrasts eflfectively 
with the delicate blue of a smaller companion, perhaps 
physically related to, though not visibly circulating round it. 

Like S Gephei, 17 Aquilse never sinks out of reach of the 
unarmed eye. At its faintest it is of 4*7, at its brightest of 
8*5 magnitude. In the mode, no less than in the range of its 
fluctuations, it closely resembles the Gepheus variable (see 
its light-curve in fig. 22). Their period of seven days, four 
hours, fourteen minutes, and four seconds, has probably 
slightly expanded since Pigott first noticed them in 1784,^ 
and is sometimes irregularly deviated from to the extent of 
several hours. The spectrum of rf Aquilse is of the solar type. 

In 10 SagittfiB, a variable discovered by Mr. Gore in 1885, 
the typical pause in descent is strongly marked. The star 

Ma«. 37 

4 5 -^ ^^^/ 

0<^1S34»678 10 

Fig. 23. — Light-curve of ( GK3zninorain. 

rises by eight-tenths of a magnitude in less than three days, 
then drops to its former level in five and a half.^ In its dim 
phases it cannot be followed with ordinary sight. 

The maximum of ? Geminorum, noticed by Schmidt in 
1847 to vary between 8-7 and 4-5 magnitudes, is, by a some- 
what rare exception, symmetrically placed as regards the 
preceding and ensuing minima (see fig. 28). The period of 
ten days, three hours, forty-three minutes, and twelve 
seconds has lengthened by ten minutes since its definition 
forty-three years ago. In one southern star the rule of an 
ascent quicker than the descent is found to be inverted. 
This is R Trianguli Australis, discovered by Gould to vary 
from 6'6 to 8 magnitude, the change upward being accom- 

> Phil Trans, vol. Ixxv. p. 127. 

« Sawyer. Astr, Jour, No. 167 ; Chandler, Astr. Nach, No. 2749. 



plished in two days, downward in less than a day and a half.' 
Another southern variable, also of Gould's detection, is re- 
markable for its excessively short period — the shortest with 
which we are acquainted except that of one Algol- variable. 
The light-curves of the two are placed in juxtaposition in 
fig. 24. B Muscae increases from 7*4 to 6*6 magnitude in 
nine hours, and subsides back again in twelve hours, twenty 
minutes, so that the total length of its cycle is only twenty- 
one hours, twenty minutes. And its changes are rendered 
conspicuous by the circumstance that, as Dr. Gould remarks, 

Mag. 6-0 



Fig. 24.— Light-oorves of U Ophiudii (No. 1), and B Muscas (No. 2), the 
two variables of shortest known periods. 

' its average brightness is so near the limit of ordinary visi- 
bility in a clear sky at Cordoba that the small, regular 
fluctuations of its light place it every few hours alternately 
within or beyond this limit.' * 

The variability of stars like yS LyrsB and 17 Aquilae is 
almost more difficult to explain than the variability of stars 
like Mira Ceti. The simspot analogy, it is true, survives in 
the similarity of the curves picturing the flow in time of both 
descriptions of change (compare figs. 16 and 21), but with 
profound differences. We are, accordingly, as good as de- 
stitute of a physical explanation of short-period variability. 
Mathematical theories on the subject have, indeed, been 

Uran, Argentina^ p. 260. 

Ibid. p. 258. 


elaborated by Gylden* and Pickering, but they scarcely 
profess to be more than devices for geometrically representing 
the progress of the phenomena. The combination is assumed 
in them of exceptional and barely possible modes of axial 
rotation with a highly elaborate arrangement of slag-covered, 
or otherwise darkened areas over the surfaces of stars, 
which, at least in Gylden's view, must be mainly solid bodies. 
But spectrum analysis shows one of them (yS Lyrae) to be of 
a gaseous constitution, and most of the others to resemble the 
sun ; so that the existence on them of permanent patches of 
obscuration is out of the question. We are often called upon, 
in natural investigations, to admit what greatly widens 
experience, but we are always justified in rejecting what 
contradicts it. 

The hypothesis of M. Plassmann ' associates bodily tides, 
causing (it is assumed) a temporary increase of luminosity as 
they travel, with occultations by the tide-raising satellites. 
So artificial a combination seems little likely to be realised ; 
yet we are reminded by Professor Pickering * that improba- 
bility must be expected in any theory of phenomena so 
antecedently improbable as those of variable stars. By Mr. 
Lockyer, on the other hand, they are considered due to recur- 
ring meteoric infalls ; and both views, we believe, will prove 
to be correct in ascribing a compound nature to the objects in 
question. Their vicissitudes represent, in some way, there is 
reason to suppose, the effects of orbital movement ; in what 
precise way, must for the present remain an open question. 
Electrical changes are no doubt concerned, but our ideas as 
to what these imply are so vague, that to appeal to them 
seems like catching at the easiest expedient for avoiding a 
confession of total ignorance. 
s/^y^ stars of the Algol type, forming Pickering's fifth class of 
variables, stand in many respects apart from all other ob- 
jects of the kind. Their peculiarities are so marked, and, 
on the whole, so rigidly maintained, as to invite close scrutiny 

* Acta. SocUtatis Fennica, t. xi. p. 345. 

' Die verdnderlichen Sterne^ p. 54. 

■ Proc. Amer, Acad, vol. viii. (1881), p. 259. 



and minute comparison with theory. Variability is in them 
by short accesses, and consists always in a temporary loss of 
light. They undergo in fact what are now known to be real 
ecHpses at stated intervals, while shining, for the most part, 
as steadily as ordinary stars. Their detection is for this 
reason so difficult that their scarcity may be more apparent 
than real. Up to the present, acquaintance has been made 
with only nine, the designations of which follow in the order 
of their discovery. 




Amount and dnntion 
of obange 


H. M. 


MAO. MAO. H. If. 

Algol . . 

MontaDari, 1669 


20 48 


2*2 to 8-7 in 10 

S Gancri . 

Hind, 184S 


11 87 


8-2 to 9-8 in 21 80 

\ Tauri . 

Baxendell. 184S 


22 50 


8-4 to 4*2 in 10 

8 Libr m . 

Schmidt, 1869 


7 61 


4-9 to 61 in 12 

U CoronaB . 

Winnecke, 1869 


10 61 


7-6 to 8-8 in 9 42 

UCephei . 

Geraski, 1880 


11 47 


7-2 to 9*4 in 10 

U Ophimhi 

Sawyer, 1881 

20 7 


60 to 6-7 in 4 80 

YCygni . 

Ghandler, 1886 


11 58 


7-1 to 7-9 in 6 

B Gania Majoris 
1 1- cv»aLAt. 

f'Jl^^ iff. 


2 43 
7 fl 


5-9 to 6 7 in 5 


The first star on the Ust is the most accurate in its changes, 
and has been the most accurately observed of any star in the 
heavens. Their extraordinary character was determined, and 
an explanation of them by interpositions of a dark sateUite 
suggested, by Goodricke in 1788, since when, some 18,600 
minima have occurred in a manner perfectly consistent with 
the hypothesis. It became then of great interest to test its 
absolute truth, and the first means of doing so were afforded 
by Professor Pickering's strict inquiry into the conditions o 
the supposed recurring ecUpse.* They proved to be in Algol 
all but perfectly complied with. Outside of the ten hours 
during which it parts with and regains three-fifths of its 
light, the star displays the required uniform lustre. The 
oscillation is of stereotyped pattern, the same in duration and 
extent now that it was fifty years ago, and that it will be 
probably fifty years hence. The precision of its performance 
seemed to correspond far better with the results of geometrical 
rule and measure than with those of the complex interaction 

^ Proc. Amer, Acad, vol. viii. (1881), p. 17 ; Observatory, vol. iv. p. 116. 


of physical causes. The spectroscope testified in the same 
sense by showing the surviving light at minimum to be 
of unchanged quality. It is dimmed, as if in large measure 
cut off, but betrays no symptom of intrinsic modification. 
These singular correspondences have not proved deceptive. 
The postulated eclipses actually take place. 

The maimer in which their genuineness has been estab- 
lished illustrates the extraordinary versatility of modern 
methods of research. No problem in which distant light- 
sources are concerned seems beyond their capacity to grapple 
with. The received explanation of Algol's changes evidently 
involved the mutual revolution, ia a period identical with 
theirs, of the eclipsed and eclipsing bodies. And since 
their orbits, to admit of a transit of the satellite over the 
primary, should lie almost edgewise to our sight, practically 
the whole of their velocity should, in the course of each 
revolution, be directed alternately straight away from, and 
straight towards the earth. Here, accordingly, spectroscopic 
measures, recommended by Professor Pickering,* were clearly 
applicable, and after being tried visually with some promise 
of success at Greenwich,^ were decisively brought to bear, 
photographically, by Professor Vogel in 1888-9.' Their re- 
sult was one of the most remarkable verifications of theory 
on record. 

Before each minimum Algol was found to be moving 
away from the sun (independently of a continuous translation 
towards him of 2^ miles a second) at the rate of 26^ English 
miles per second; after each minimum, to be approaching 
with an equal speed; while at iatermediate times the im- 
printed lines, by resuming their normal positions in the 
spectrum, showed the star to be then moving perpendicularly 
to the visual ray. Multiplying this velocity (of 26J miles) by 
the number of seconds in Algol's period (247,732) we get 
an orbital circumference corresponding to a diameter of (in 
round numbers) two million miles. Moreover, since the pro- 
portionate dimensions of the bright and dark bodies are shown 

» Proc, Amer. Acad, vol. viii. p. 34. « Observatory^ yol. xi. p. 108. 

• Astr. Nach. No. 2947. 



by the amount of obscuration of one by the other to be very 
nearly as 100 to 88, their relative masses would also be known, 
if we could be sure that they are of the same mean density. 
The assumption as regards a mass shining with great 
brilliancy and one almost totally dark is certainly a hazard- 
ous one, but it receives some warrant from the example of the 
sun and Jupiter. By its aid Professor Vogel arrived at the 
following provisional data for the system of Algol. 

Diameter of Algol 

„ satellite . 
Distance from centre to centre 
Orbital velocity of Algol 

„ H satellite . 

Mass of Algol 
„ satellite . 

1,061,000 English miles 

830,300 „ 


26-3 miles per sec. 

J solar mass 


5 »♦ 

In the accompanying diagram G marks the centre of 
gravity round which both stars revolve with velocities in- 
versely proportional to their masses. Thus, Algol travels 

in an orbit of only half the 
compass of that of its com- 
panion, because possessed 
of twice its attractive force. 
It is easy to see, too, that 
the duration of the ecUpse 
compared with the length 
of the period gives the rela- 
tion between the diameter 
of the occulted body and the 
diameter of the orbit of the 
occulting body ; whence the 
absolute dimensions of one 
becoming known, those of 
the other follow. 
The density alike of Algol and of its satellite is less 
than a quarter that of the sun, or 0*88 that of water. They 
must both then be completely gaseous, and Professor Vogel 
finds evidence, in the early and later phases of eclipse, of the 
possession by each of very profound atmospheres. Spectro- 

Fio. 25.— Algol during an Eclipse. 


seopic confirmation, however, is totally wanting to the view 
that any part of the darkening is due to atmospheric inter- 
position. A slight want of symmetry in the curve represent- 
ing the light changes before and after minimum is associated 
by Dr. Wilsing^ with the ellipticity of the satellite's path, 
causing it to slacken its course in moving off the disc. The 
deviation nevertheless from circularity must be of trifling 
amount, since otherwise the two bodies would come into con- 
tact at their nearest approach. As it is, the stability of the 
extraordinarily close system formed by them comes only just 
within the range of theoretical possibility. Although the 
period of Algol is now six seconds shorter than it was in 
Goodricke's time, and is still diminishing, it is nearly certain 
that compensation will eventually take place. The perturba- 
tion gives us perhaps the only hint that is ever likely to reach 
us of the association with Algol of further unseen com- 

It needs no argument to prove that the eclipse-theory of 
the variable in the head of Medusa must apply to all other 
members of the same sharply characterised class. Many of 
them, however, present anomalies which are the more de- 
serving of careful study that they may one day throw an 
important light on the circumstances under which combina- 
tions of the indicated kind exist. 

The light-change of 8 Gancri, the second of the Algol 
variables, was discovered by Mr. Hind in 1848, and its peculiar 
nature ascertained by Argelander in 1852.^ The star remains 
steady during thirteen-fourteenths of its period, then declines, 
in eight hours and a half, to less than one quarter of its usual 
brightness, which it recovers in the course of thirteen hours 
more. Besides this wide inequality in the times of sinking 
and rising, the latter process is interrupted soon after it has 
begun by a marked pause,' represented graphically from 
Schonfeld's obserTations in fig. 26. A large irregidarity, 
besides, has once been detected in the compass of this star's 
change. On April 14, 1882, Schmidt observed at Athens a 

» Astr. Nach. No. 2960. » IHd. Nos. 796, 804, 806. 

■ Vierteyahrsschrift Astr. Ges, Bd. ix. p. 230. 


minimum nearly two magnitudes fainter than any he had seen 
before. During one hour the star remained sunken nearly 
to the twelfth magnitude.* The period of S Gancri is subject 
to a perturbation with a range of about forty minutes, and 
embracing rather more than three hundred minima.' 

Inequalities of this kind, which in Algol sum up to a few 
seconds in a century, and grow to many minutes in S Gancri, 
are in X Tauri counted by hours.' The task is a formidable 
one of explaining them on gravitational principles in a system 
of which the average period is under four days. The same 
star affords another example of an accelerated decrease, as 
compared with the increase of its Ught, which is also believed to 

Mag. 8-S • 

Oh S 4 6 8 10 13 14 16 18 SO 33 

Fio. 26.— Minimum of S Cancri. 

be slightly variable outside the regular accesses of change.^ 
But although never descending much below the fourth magni- 
tude, it has been comparatively little observed. 

The period (about 2^ days) of S LibrsB is shorter, while 
its time of oscillation is two hours longer than that of Algol. 
It follows that the supposed eclipsing body circulates in still 
closer contiguity to its primary than the Algol-satellite. The 
gap of space from surface to surface can not, in fact, much 
exceed one -third of the interval from centre to centre. The 
mean density of the components, too, must be so far below 

» Astr. Nach, No. 2491. 

' Argelander, Bonner Beob, Bd. viii. p. 397 ; Schonfeld, Sirius, Bd. x. p. 66. 

* Schonfeld, Jahresbericht, Mannheim, Bd. xl. p. 76. 

* Plassmanni Die verdnderlichen Sterne^ p. 42, 


that of Algol/ that their subsistence as globes under the dis- 
ruptive strain of their mutual gravitation might well appear 
incredible. There is even a further aggravating circumstance. 
The recurring drop of the star from 4*9 to 6'1 magnitude, 
occupies 5^, the compensatory rise 6^ hours; so that a 
circular orbit seems precluded, while a sensibly eccentric 
one would infallibly bring about an immediate collision! 
Inequalities of 6^ seconds, comprised within a cycle of 
about nine years,^ modify the periodicity of this interesting 

The fifth Algol-variable, U CoronaB, declines from 7*6 to 
8*8 magnitude in 4h. 80m. ; regaining its normal brightness 
in 5h. 12m.; so here again we meet the relatively slow 
resumption of brightness prevalent in this class of objects. 
U CoronaB appears to be slightly variable independently of its 
systematic changes.^ The disturbances of its period have 
been studied by Mr. Chandler.* 

The variations of U Cephei, first recognised by M. Ceraski 
at Moscow, June 28, 1880, are more rapid and extensive than 
those of any of its congeners. In four and a half hours, the 
star is reduced to about one-ninth its ordinary lustre, losing 
light, at one stage of its decline, at the astonishing rate of 
more than one magnitude an hour I The obscurity lasts an 
hour and a half, but not with entire uniformity. The lowest 
point is touched at first,^ and a pause in the ascent, like that 
inflecting the light-curve of S Cancri (see fig. 26), is indicated. 
Some complicated irregularities of period have further been 
ascertained by Mr. Chandler. The white rays of U Cephei 
turn ruddy at minimum, or rather, if we adopt Professor 
Pickering's suggestion,® the light of an enormous but im- 
perfectly luminous satellite is substituted for that of the star. 
For the eclipse is in this case supposed to be total during the 
entire ' stationary period ' of an hour and a half; was supposed 
we should say, for Mr. Chandler's observations of a checked 

1 See Mr. MazweU Hall's computation as regards Algol, Observatory, yoI. ix. 
p. 225. 

« Schdnfeld, Jahresbericht, Bd. xl. p. 96. • Ibid, p. 96. 

* Chandler, Astr, Jour. No. 206. » Ibid, No. 199, p. 63. 

• Proc, Amer, Acad. vol. viii. (1881), p. 389. 


attempt at increase during the time assigned to totality are, 
if confirmed, inconsistent mth its prolongation. 

A star of the sixth magnitude in Ophiuchus, detected as 
an Algol-variable by Mr. Edwin Sawyer, of Cambridge-pprt, 
Massachusetts,^ had its true period of twenty hours, seven 
minutes, and forty-two seconds, the shortest yet met with in 
any star, fixed by Mr. S. G. Chandler.' The obscuration lasts 
four and a half hours, and is about equally divided between 
a fall and a rise of seven-tenths of a magnitude.' But the 
return of light does not proceed uniformly (see fig. 24, No. 1). 
It is arrested, about half an hour after minimum, by a * stand- 
still' of some fifteen minutes, independently recorded by 
Sawyer and Chandler.^ Moreover, the satellite producing this 
anomalous eclipse must, it would seem, revolve quite close to, 
if not actually within the limit of distance shown by Boche 
of Montpellier to be the least at which an attendant globe 
could maintain its integrity against the tremendous strain of 
tidal forces. The average density of these conjoined bodies 
comes out less than one-fifth that of the sun, or 0*27 that of 
water. A remarkable observation of U Ophiuchi was made 
by Schjellerup, June 9, 1863.* He noted it with surprise 
(being aware that other observers had seen it much brighter) 
as of 7'7 magnitude, or one full magnitude below the lowest 
it has been known to touch in recent years. 

The period of T Gygni, added to the list of Algol-variables 
by Mr. Chandler, December 9, 1886, averages about a day 
and a half,^ but fluctuates to an extent unparalleled in this 
kind of star.^ The retardation of its phases between 1887 
and 1888 amounted to seven hours, totally disconcerting pre- 
diction, and the period is now shortening as rapidly as it 
lengthened before.® The actual change does not exceed half 
a second at each of the returns, but these are so numerous, 
that the accumulating errors sum up in a short time to a 
startling aggregate. The deviations of T Cygni, in fact (as 

» Astr. Nach. No. 2412. « Astr, Jour. No. 161. 

« Astr. Nach. No. 24S4. * Astr. Jour. Nob. 162, 177. 

* Catalogue of 10,000 Stars, Copenhagen, 1864, No. 6162, and note. 

• Chandler, Astr. Jour. No. 163. 

» Chandler, ibid. No. 186. - Ibid. No. 204. 



Mr. Chandler points out), exceed those of Algol a hundred, 
those of U Ophiuchi a thousand fold. 

The twenty-seven hours' period of R Canis Majoris is 
divided between apparently symmetrical oscillations lasting 
five hours, and a steady maximum of twenty-two hours' dura- 
tion. Its individual peculiarities have yet to be determined ; 
but the conditions of change are visibly such as can only be 
realised in a surprisingly close system. 

Our survey of the Algol-variables has thus disclosed the 
following significant peculiarities. 1. In six out of nine 
cases, the eclipse deepens more rapidly than it lightens ; in 
no instance is this relation inverted. 2. All the systems con- 
sidered are made up of bodies much more tenuous than 
water, one member of each pair being nevertheless sensibly 
obscure,^ while the other is brilliantly luminous. 8. All (so 

Ma«. 6-4 

0A1S34 ft 6 78 

Fia. 27. -Inverted Light-ourre of 10 Sagitts (Gbre 1885). 

far as is known) are subject to periodical disturbances from 
invisible attracting masses. 4. Two of these stars (S Cancri 
and U Ophiuchi) have been observed to undergo at minimum 
a loss of light far exceeding the usual proportion. 5. Several 
Algol-variables invert most curiously the typical light-curves 
of such stars as S Cephei and rj AquilsB (figs. 21, 22). To 
the quick advance of darkening in the Algol-stars corresponds 
a quick gain of light in B Cephei and its allies ; the pause in 
the ascent of the one class is represented with strange fidelity 

1 Vogel calculates that the companion of Algol must be, at any rate, eighty 
times less luminous ; since otherwise a second minimum would be perceptible 
corresponding to its occultation by the primary star. 


by a halt in the decline of the other. We give in fig. 27 the 
reversed light-curve of 10 Sagittffl, discovered by Mr. Gore 
in 1886 as variable in 8j^ days. The resemblance to that of 
8 Gancri (fig. 26) is unmistakable. We seem to have before 
us the characteristic minimum of an Algol star. It is di£S- 
cult to believe in the total dissimilarity of the causes pro- 
ducing effects opposite indeed, yet displaying in their oppo- 
siteness so remarkable an analogy. The additional remark 
may be worth making that ' eclipse-stars * seem to agree in 
showing a Sirian spectrum; unless indeed B Ganis Minoris 
prove an exception to the rule.^ The solar type, on the 
other hand, predominates among ordinary short-period vari- 
ables. Variable stars of all classes are probably at enormous 
distances—even on the celestial scale— from the earth. There 
is no sign that any of them are included among the stars in 
our comparative vicinity. One of the best means of form- 
ing a rough general judgment on this point is by amount 
of apparent motion ; and variables remain in general nearly 
fixed in the sky. Mira, one of the most mobile, shifts its 
position indeed to the not wholly inconsiderable extent of 
twenty-five seconds of arc in a century; but measures for 
parallax would be much embarrassed by its changes of magni- 
tude, and have not yet been attempted. We are thus abso- 
lutely without direct information as to the remoteness of 
variable stars. 

Their distribution over the sphere presents some noticeable 
peculiarities. Contrary to what might have been expected, 
short-period variables, although on the whole much brighter 
objects than those of long periods, tend much more decidedly 
to concentration in the Milky Way, while the preference for 
its plane among those not variable belongs chiefly to faint 
stars. In Algol-stars, though present, it is less strong than in 
periodical stars of Pickering's fourth class. These lie, for the 
most part, along a great circle nearly, but not quite coincident 
with the medial line of the galaxy.^ It is remarkable that 

> See Pickering, Henry Draper Memorial, Fourth Annual Report^ p. 6. 
* The northern pole of this circle, according to Pickering, is situated in 
R.A. 13h., Deo. + 20^ That of the Milky Way is in R.A. 12h. 40m., Deo. + 28°. 


their condensation-level (as is shown by its being projected 
into a great circle) passes through the sun. Within the zone 
itself, there is an evident disposition towards clustering. 
Where the Milky Way divides in Cygnus, the variables follow 
its southern branch, and they are thickly sown over the whole 
sky-region from Lyra to Sagittarius.^ Indications, indeed, 
abound that the conditions of variability, and even of par- 
ticular kinds of variability, are localised in space. Thus in 
Sagittarius, no less than four stars fluctuate in periods of six 
to seven days, and several others are subject to slower vicissi- 
tudes. Two adjacent stars in the Southern Triangle vary in 
unusually short periods. E follows S Coronae Australis 
nearly in the same parallel, after an interval of only forty-six 
seconds of time. The new star which appeared in Scorpio in 
1860 marked the centre of a group of six objects, all widely 
variable irregularly or in long periods. Five stars of a similar 
nature, including two virtually extinct Novsb, are collected 
in a small section of Ophiuchus ; and in general the sites of 
temporary stellar apparitions are more or less closely dotted 
round with variables. There is reason to suppose that the 
circumstances favouring instability of light do not exist any- 
where in the neighbourhood of the sun. 

* Chandler, Astr. Jour. No. 193 ; Plassmann, Die verdnderlich&n Sterne, 
p. 85. 




The stars differ obviously in colour. Three or four among the 
brightest strike the eye by their ardent glow, others are tinged 
with yellow, and the white light of several has a bluish gleam 
like that of polished steel. Reddish tints are, however, 
in the few cases in which they affect lucid stars, the most 
noticeable, and were the only ones noticed l)y the ancients. 

Ptolemy designates as * fiery red ' {viroKLppoC) the following 
six stars: Aldebaran, Arcturus, Betelgeux, Antares, Pollux, and 
— mirahile dictu — Sirius ! all the rest being indiscriminately 
classed as 'yellow ' {^dvdoi). Now Pollux at present, though 
by no means red, is at least yellowish, but Sirius is undenia- 
bly white with a cast of blue. A marked change in its colour 
since the Alexandrian epoch might thus at first sight appear 
certain, the more so that Seneca makes express men- 
tion of the dog-star as being * redder than Mars ; ' * Horace 
has * rubra Canicula ' as typical of the heat of summer ; * and 
Cicero, in his translation of Aratus, speaks of its 'ruddy 
light.' Nevertheless the case is doubtful. The questionable 
epithet, in all probability, crept into the * Almagest ' by a 
transcriber's error, Ptolemy not being responsible for it. In 
the early Arabic versions of that work it evidently did not occur, 
for Arab astronomers of the tenth and subsequent centuries 
ignored the imputation of colour to the dog-star, and Albategnius 
stated the number of Ptolemy's red stars as five.^ Among the 
Latin writers, the misapprehension (if misapprehension there 
were) originated with Cicero, who was much more a rhetori- 

» Qiurst. Nat. I. i. « Sat. ii. 6, 39. 

' W. T. Lynn, Observatory, vol. x. p. 104. 


Clan than a natural philosopher, and it became practically 
extinct with the perhaps unverified assertion of Seneca. 
It is, however, curious to find * the fiery Sirius ' coming up 
again in the verses of so close and original an observer as 

There is better reason to believe that Algol has either 
really blanched with time, or may be subject "to temporary 
suffusions of colour. It is now purely white, with a 
spectrum of the Sirian pattern, but appeared red in the 
tenth century to the Persian astronomer Al Sufi.* His 
authority is considerable, and his only other addition to the 
ruddy stars of Ptolemy is fully justified by the ardent glow 
of *Cor Hydrae.* It is worth recording, too, that Schmidt 
noticed in 1841 the ' demon-star ' in Perseus as yellowish 
red, although he never in later years saw it otherwise than 

The same observer was amazed, March 21, 1852, to per- 
ceive Arcturus without a trace of the strong colour familiar 
to him in it during eleven previous years. In comparison 
with its paleness, Capella seemed bright yellow. Mars and 
Betelgeux glowed almost like fire.^ It was some years before 
the star resumed its original hue, and the reality of the 
change, admitted by Argelander, was certified by the observa- 
tions of Kaiser at Leyden.^ 

The periodical variations in colour of a UrssB Majoris, 
the 'Pointer' next the pole, announced by Klein in 1867,'* 
were long disbelieved in. Yet they appear actually to take 
place, though perhaps with much fitful irregularity. A series 
of observations with Zollner's polarising colorimeter, executed 
in 1881 by M. Kovesligethy, of the Gyalla observatory in 
Hungary, gave clear evidence of alternating fluctuations 
between red and yellow in a period of 54^ days.® The star 
is under no suspicion of varying in light. 

The colours of the stars visible to the naked eye are faint 
and pale compared with those disclosed by the telescope. The 

' Schjellemp, Description des £toiles FixeSy p. 26. 

* Astr. Nach. No. 1099. » Ibid. No. 999. 

* De Stcrrpnhfmel verUuart, Pt. i. p. 597. 

» Astr. Nach. Nos. 1663, 2111. • Sirius, Bd. xiv. p. 2.'J3. 

1. 2 


real gems of the sky are found low down in the scale of bright- 
ness. To some extent this is only what might be expected. 
Intense tints result from strong selective absorption in the 
atmospheres of the stars they distinguish, and strong 
absorption implies large loss of light. Stars shine with the 
rays that have survived transmission through the glowing 
vapours in their neighbourhood, and the more nearly those 
rays are limited to one particular part of the spectrum, the 
purer and clearer the resulting tint will be. A true prismatic 
hue could accordingly be produced only through an enormous 
reduction of brightness ; but true prismatic hues do not exist 
among the stars,* the colours of which are always more or 
less copiously diluted with white hght. 

The science of star-colours has hitherto made little satis- 
factory progress. Attempts to set up a standard chromatic 
scale have not been successful,* and instrumental devices for 
ensuring just and equable judgments may sometimes induce 
larger errors than they avert.^ Simple visual estimations, on 
the other hand, must be treated with great reserve, since 
* personal equation ' in this matter often assumes enormous 
proportions. The extreme of colour-blindness is reached by 
comparatively few, but endless minor individualities of per- 
ception vitiate the greater part of an accumulated mass of 
evidence which might otherwise justify inferences of real 
change. From the complex bundle of rays forming the 
image of a star, each retina picks out and accentuates those 
to which it is most highly sensitive, precluding the possibility 
of agreement as to delicate tints between many different 
observers. With both the Herschels, for instance, the equili- 
brium of colour was shifted towards the red end of the 
spectrum. Sir William's * garnet ' star (ji Cephei), Sir John's 
' ruby ' stars, have appeared merely reddish yellow to sub- 
sequent observers. Struve's assistant, Knorre, saw all stars 
indiscriminately white; Admiral Smyth, on the contrary, 

* Struve, Mensurts MicrometriccRy p. Ixxxvi. 

* See the system proposed by Franks, Monthly Notices, vol. xlvii. p. 269. 

' See the results given by Kovesligethy, Ueher eine neue Methode de 
Farhenbestiminung dcr StemCy Halle, 1887. 


discriminated between shades of colour altogether inappreciable 
to most of those who have profited by his * Cycle of Celestial 

Even of the same observer the impressions do not always^ 
agree. Fatigue and advancing years modify the colour- 
sense ; and M. Safarik states that stars invariably appear 
redder to his left than to his right eye.* Atmospheric condi- 
tions, too, are powerfully operative. Misty air blots out faint 
tints and alters strong ones, azure visibly turning green 
through its mfluence. Height above the horizon is another 
circumstance to be taken into account before any useful com- 
parisons can be made, while instrumental causes tend further 
to perplex their upshot. Large apertures in themselves help 
to bring out colour, especially in small stars ; but the colour- 
correction of great refractors is always imperfect, and the out- 
standing blue fringe usually conspicuous in them must by 
contrast give a reddish tone to the image.^ Eeflectors produce 
a similar effect through absorption of many of the higher rays 
by the silvered glass or speculum metal forming their mirrors. 
Indeed, Mr. Franks, an assiduous cultivator of stellar chro- 
matics, thinks there is not much to choose in this respect 
between reflecting and perfectly achromatic telescopes,^ al- 
though their size and adjustment make a very great difference. 
Above all, only medium powers should be used in colour 
observations, since with high magnification all hues merge, 
or tend to merge, into yellow. 

In this department, then, there is often little connection 
between appearances and realities. Discrepant statements are 
far from necessarily implying actual variation. The former 
abound ; instances of the latter are to be met with, but can 
only be admitted with extreme caution. 

The study of star-colours divides naturally into two 
branches — one concerned with isolated, the other with com- 
pound objects. Inquiries in the first case are simplified by 

* Vierteljahrsschrift Astr. Ges. Jahrg. xiv. p. 378. 

» Webb, Student, vol. v. p. 487. 

' Jour. Liv. Astr. Soc. vol. v. p. 88. Mr. Franks considerB that, on the 
whole, colour pales with increase of light. Struve held the opposite opinion. 
Bee Mc7is. Microm. p. Ixxxvi. 


the curious and unexplained fact that single stars are never 
markedly tinged except with red or yellow. Vega makes the 
nearest approach in the northern hemisphere to an indepen- 
dently blue star ; 7 Toucanae, a Eridani, and e Pavonis are 
the * pale sapphires ' of the southern sky. But they are very 
pale indeed, so pale as to produce no definite impression of 
colour upon ordinary eyes. Nor is the * emerald' tinge of 
P Librae more decided. We have accordingly to deal just at 
present only with * red stars.' 

The earliest list of thirty -three of them was drawn up by 
Lalande in 1805.* * Ces etoiles,' Von Zach remarked in re- 
publishing it in 1822, * annoncent toujours quelque chose de 
particulier; or toute particularite merite d'etre observee.'^ 
We have to a great extent got rid of the notion which pre- 
sented itself to John Michell m 1767,^ that what they * an- 
nounce ' is the impending extinction of their own fires, but 
their peculiarities have become, on that very account, all the 
more worthy of attention. Red stars are commonly variable 
both in light and colour ; the display of colonnaded and zoned 
spectra belongs exclusively to them; and they are strongly 
characterised by atmospheric extent and instability. Few of 
them can be watched long and attentively without being 
caught in some singular phase of change. 

Their systematic study b^gan with the publication in 1866 
of the late M. Schjellerup's Catalogue of 280 red stars ; ^ ten 
years later, Mr. Birmingham of Tuam completed a similar 
work comprising 658 entries,^ and Mr. Chambers laid before 
the Eoyal Astronomical Society, April 6, 1887, a catalogue 
founded on his personal observations during seventeen years.® 
Of 711 nominally * red ' stars in both hemispheres, he had 
examined 589, being virtually all those visible in England, 
with the result of finding the colour of most exaggerated. 
* Orange ' was to his eye the tint prevailing among them ; 

' Connoissance des Tents pour Van 1806. 

2 Corresp. Asir. t. vii. p. 296. ■ Phil. Trans, vol. Ivii. p. 238. 

* Astr. Nach. No. 1591 ; reprinted with numerous additions in Vierteljahrs- 
schrift Astr. Ges. Jahrg. ix. p. 252. 

* Memoirs R. Irish Acad. vol. xxvi. p. 240. 

* Monthly Notices^ vol. xlvii. p. 318. 


true *red8 ' were scarce; of stars meriting to be qualified as 

* carmine ' or * ruby * he had not met above a dozen. 

Observers in this line have indeed usually expressed dis- 
appointment with the tints described by their predecessors. 
Sir John Hersuhel's seventy-six * ruby-coloured objects,' * fall 
short, by general admission, of the expectations raised by that 
epithet. Schjellerup found most of Schmidt's red stars nearly 
colourless, Birmingham made the same remark as to many 
of Schjellerup's, Chambers is dissatisfied with Birmingham's. 
Bii-mingham did not, however, claim much depth of colour for 
most of the objects he enumerated ; he fully admitted their 

* redness ' to be, with few exceptions, largely adulterated with 
orange, and Schmidt has pointed out that there is absolutely 
no star of the pure and perfect crimson of a solar prominence. 
This is necessarily the case, since no stellar atmosphere is 
entirely opaque to rays in the upper part of the spectrum. 

It would, nevertheless, be a mistake to suppose that red 
stars are not striking telescopic objects. Their light, even in 
the less distinguished specimens, has a lurid glow which at 
once marks them out from ordinary stars, and those of 
deeper tmts shine with an ardour recalling the wrathful 
intensity of a stormy sunset. The contrast between a red 
and a white star in the same field of view is sometimes most 
beautiful and effective. Thus, in the southern constellation 
Grus, TT* and tt^ show like little burnished discs of copper and 
silver respectively, seen under strong illumination. 

Among conspicuous stars, Antares, in the heart of the 
Scorpion, is the ruddiest, Betelgeux comes next, and both are 
suspected to vary in tint. Aldebaran and Arcturus have . 
figured immemorially in the short list of visibly fiery objects, 
to which Al Sufi added a Hydrae, and Father Noel 7 Crucis. 
But their colours are mere pale shades compared with those 
instrumentally brought into notice. * Hind's crimson star,' 
otherwise known as * E Leporis,' appeared to its discoverer in 
1845 like ' a drop of blood on a black field.* As with most 
other variables, however, increase of light brings with it a 
paling of colour, intense redness giving place to a coppery 

' Cape Observations, p. 448. 


hue as it rises above the eighth magnitude. Its spectrum is 
of the fourth type, with particularly strong absorption of the 
blue rays, a very small proportion of which penetrate the veil 
of dense vapours surrounding this remarkable object. 

A similar one, now known as V Hydrse, is No. 16 of 
La'.ande's, No. 136 of Schjellerup's Eed Stars, and was re- 
corded by Dr. Copeland at Dunsink, March 22, 1876, as 
' brown red/ and of 7*2 magnitude.' But three years later, 
Dr. Dreyer found it risen to the sixth magnitude, and of a 
* most magnificent copper red,' while Birmingham observed it 
in 1874 as of the eighth, Duner in 1884, down to 9*5 magnitude. 
Its fluctuations of light are thus very considerable. 

Close to one of the gems of the Southern Gross, an eighth 
magnitude star was described by Sir John Herschel as of * the 
fullest and deepest maroon-red, the most intense blood-red of 
any star I have seen. It is like a drop of blood when con- 
trasted with the whiteness of fi Crucis.'^ Among other 
southern stars remarkable for colour, are B Sculptoris, 
no less ' intensely scarlet ' now than when Gould saw it at 
Cordoba; E Doradus, glowing like a live coal out of the 
darkness of space, and Lg Puppis; all of them noted va- 

In the northern hemisphere, V Cygni bears the palm for 
depth of tint, which, however, pales somewhat towards its 
maxima; and not far inferior to it are E Cassiopeiae, B 
Leonis, B Crateris, and Mira, with U Cygni and U Cassiopei®, 
both splendidly set off by the vicinity of blue attendants. 
(Crimson, indeed, verges, in these and other periodical stars, 
more and more towards orange in their brightening phases ; yet 
they always remain technically ' red.' A few cases of complete, 
if temporary change of colour have, however, been recorded. 
Thus, a seventh magnitude star in the Lynx (90 Schjellerup) 
noted by Struve as * rubra,' by Secchi as * bella gialla,' seemed 
blue, or bluish white to Birmingham January 13, 1874, a con- 
firmatory and nearly contemporaneous observation being made 
at Greenwich.' A star of 8^ magnitude (148 Schjellerup) 

* Duiisink Observatums, vol. iv. p. 55. * Cape Observations^ p. -148. 

■ Memoirs R. Irish Acad. vol. xxvi. p. 209. 



called ' scarlet ' by Lord Eosse in 1861, * dark red ' by d'Arrest 
December 8, 1866, showed no colour to Birmingham May 8, 
1874. Duner found it, nevertheless, of a deep orange red in 
1884, and it is characterised by a fine colonnaded spectrum. 
Again, Schjellerup was struck with the redness of a star in 
Aquila* in 1868, which after an interval of ten years, ap- 
peared to Birmingham actually blue ; and similarly, a bluish- 
white object occupied the place, November 14, 1850, of a star 
in Taurus marked ' very red ' by Mr. Hind, September 3, 1848.^ 
One further instance may be mentioned, in the hope of at- 
tracting the attention of southern observers to an object of 
particular interest. A fifth magnitude star in Argo, known as 
r Velorum,^ was notoriously red during Gould's stay at Cor- 
doba. But it seemed to the present writer at the Cape in the 
autumn of 1888, perfectly white, or tinged, if tinged at all, 
with blue; and a spectrum to correspond, continuous, save 
for a probable dark F, was derived from it. We have unluckily 
no knowledge as to the nature of its spectrum in its antecedent 
red stage. 

The following list collects a few of the best authenticated 
ex8.mples of colour-change among the red stars. 

: Name 



1 Schjellerup 


Buby to slight tinge of red. 

4 Schjellerup 


Deep to pale red. 

5 Schjellerup 

7-8 (var.?) 

Garnet to white. 

8 Schjellerup 


Very red to white. 

Copenhagen Catalogue 3282 


Red to blue. 

63 Schjellerup . 


Red to bluish white. 

90 Schjellerup . 


Red to bluish white. 

148 Schjellerup . 


Scarlet to white. 

168 Schjellerup . 


Orange to white. 

214 Schjellerup . 


Red to blue. 

222»» Schjellerup. 


Red to white. 

The changes of colour visible in temporary stars have 
generally been in an opposite direction to those of ordinary 

* No. 6803 of the Copenhagen Catalogue = "^0, 214 of Schjellerup's Red 

2 Monthly Notices^ vol. xi. p. 46. 

» The place for 1876 of the star is R.A. lOh. 16m. 48s., D.-41° 1*3'. 
Tliere is another r Argus of the same magnitude. 


variables. Their sanguine tints faded, instead of deepening 
with the decline of their ligfit. Thus, Tycho's star, though 
it passed through an intermediate stage of redness, was of a 
leaden white when it disappeared. T Coronge ran nearly the 
same course. Nova Ophiuchi (1848) and Nova Andromedje 
were ruddy at first, colourless later. Nova Cygni from orange 
turned bluish. The corresponding colour phases of a star a 
little to the north of a nebula in Leo (New Gen. Cat., 3107) 
seem to mark it as of peculiar constitution. It fluctuates from 
about eighth to ninth magnitude, and simultaneously from 
decided orange to slight blue.' One of the very few periodical 
stars which lose colour and light together is T Geminorum. 
Below the ninth magnitude it has scarcely a tinge of yellow ; 
at maximum it flashes deep orange or red.^ 

Eed stars are very unequally distributed. Certain wide 
tracts of the sky are nearly destitute of them ; in some, they 
occur profusely. The Milky Way between Aquila, Lyra, and 
Cygnus was called by Birmingham the * Bed Eegion ' ; ' but 
other galactic constellations, such as Perseus and Cassiopeia, 
are all but exclusively formed of white stars.* 

Evidently, however, real partialities of colour -distribution 
must be to a great extent masked by the projection, to the 
eye, of objects at indefinite distances from each other upon 
the same portion of the sphere. Hence extensive local 
collections of similar stars may be so confused with overlying 
and underlying aggregations as to be completely unrecognis- 
able. Smaller groupings are more readily detected. It is by 
no accident that, in the immedia^ neighbourhood of one red 
star, others are so apt to be met with ; and the * brick red ' 
and ' ruby ' pairs included in Herschers Cape list, may w^ith 
confidence be assumed to be severally in some sort of physical 
connection. Eed stars, it was remarked by the same authority, 
are conspicuous in many clusters both by brightness and 
situation; and Father Secchi was struck with the critical 

' Dunsink Observations, vol. iv. p. 54. 
« Hind, AstT. Nach. No. 839. 
» Memoirs R. Irish Acad. vol. xxvi. p. 255. 

* Franks, Monthly Notices, vol. xlvi. y. 343 ; see also Ostboff, WochcJiscJn ijt 
filr Astr. Bd. xix. (IHTO), p. 320. 


positions of such objects as regards spiral or radiated stellar 
arrangements in the Milky Way.^ 

The principle of colour by association is barely indicated 
in clusters, and chiefly in a * jewelled ' one about k Crucis, as 
to which we shall say a few words by-and-by.^ It is in 
double stars carried out to the highest perfection. Nature is 
inexhaustible in her display among them of harmonies, con- 
trasts, and delicate gradations of hue. They not only vividly 
sparkle in green and gold, azure and crimson, but shine in the 
sober radiance of fawn and olive, lilac, deep purple, and ashen 
gray. Chalcedony, aquamarine, chrysolite, agate, and onyx 
have counterparts in the heavens as well as rubies and emeralds, 
sards, sapphires, and topazes. These beautiful tints do not 
occur at random. We can partially discern some * law of 
order ' governing their development ', but empirically as yet, 
and without any true insight into its cause. 

Mariotte of Dijon, a physicist, but no astronomer, was the 
first to speak of hlue stars. * Les etoiles qui paraissent bleues,' 
he wrote in 1681, ' ont une lumiere faible, mais pure et sans 
melange d'exhalaisons.'^ But he gave no examples, and it is 
not easy to divine what class of objects he alluded to. The 
chromatic observation of double stars was really begun by 
Father Christian Mayer, at Mannheim in 1776 ; the interest 
of his preliminary efforts was, however, completely absorbed 
in the splendour of Herschers similar, but vastly more ex- 
tensive and assured results. He not only discovered a great 
number of exquisitely tinted couples, but by his success em- 
phasised the importance o(isystematic attention to colour in 
double stars. 

His example was followed by F. G. W. Struve, who in 1887 
classified from this point of view 596 of the brightest known 
stellar pairs. The upshot was to prove agreement in colour 
the rule, contrast the exception.* Just half, or 295 of the 
objects examined had both components white ; 118 had both 
yellow or reddish with slight differences of intensity ; sixty- 

* Atti dci Nuovi Lined, t. vii. p. 72. * See i7ifr. p. 240. 

* Oeuvrcs, t. i. p. 287. 

* Etoiles Doubles, pp. 33-4 ; Mciisura Microm. p. Ixxxii. 


three were tinged with blue. The instances of genuine con- 
trast numbered 120, and in all of these the small star was 
called blue. The rule is, moreover, without exception, that 
no primary member of a dissimilarly tinted pair is blue. 

The reality of chromatic contrasts in double stars was 
established by the persistence of colour in satellites during the 
obliteration of their primaries by an interposed wire or bar ; 
and besides, as Struve remarked, optically produced tints 
should be invariably complementary, which is far from being 
the case. A curious proof of this independence was afforded 
by a double occupation of Antares and its companion, observed 
by Dawes in 1856. The small star, emerging first from 
behind the moon, seemed as perfectly green viewed thus alone, 
as when half lost in the glare of the great red star it is 
attached to.* The same phenomenon was re-observed in 

The connection between inequality of brightness and in- 
equality of colour in coupled stars did not escape Struve's 
notice.^ He found a mean difference of less than half a 
magnitude between the exactly similar members of 375 pairs ; 
of over one magnitude for 101 stars showing varied shades of 
the same colour ; and of nearly two magnitudes in 120 cases 
of contrasted tints. Professor Holden, taking into account 
all known physical pairs and none others, reached, in 1880, 
an analogous result.^ Where there was identity of colour 
the average difference of lustre proved to be only half a mag- 
nitude; where there was diversity, the luminous inequaUty 
mounted to two and a half magnitudes. One hundred and 
twenty-two of the stars considered belonged to the first class, 
forty to the second. Now markedly unequal are generally 
wide pairs,* so that disparity of hue is seen to prevail in 
systems formed by a large star and a comparatively small 
and remote companion ; while genuine twin suns, of not very 
different radiative power, and of similar radiative quality, 

• Monthly Notices^ vol. xvi. p. 143 ; Niesten, Ciel et Terre, t. ii. p. 96 ; 
Webb, Cel, Objects, p. 386. 

" Mens. Microm. p. Ixxxii. • AtJier, Jour, of Science, vol. xix. p. 467. 

* Doberck, Aslr. Nach. No. 2278. 


circulate as a rule rapidly and closely round their common 
centre of gravity. Why this is so we cannot tell ; the bare 
fact is before us. 

A good many beautifully coloured stars are, nevertheless, 
ascertained to be in mutual revolution. The yellow and 
rose-red components of ?; CassiopeisB finish their circuit in 
about two hundred years ; those of e Bootis, shining chrome 
yellow and sea-water blue, in probably upwards of twelve 
hundred ; f Bootis and tt Cephei, orange and purple, o Cephei 
and T Cygni, golden and azure pairs, are all in undoubted 
orbital movement. A good many richly tinted stars, on the 
other hand, appear stationary, doubtless because their dis- 
tances apart are so considerable as to make their revolutions 
inordinately slow. Thus the emerald-green companion of 
a Herculis has preserved, during a century, an invariable 
position with regard to the ruddy star it depends upon ; and 
Antares forms with its sea-green satellite a somewhat similar 
and equally rigid combination. The fixed pair fi Cygni 
(Albireo) shining with ' yellow topaz ' and * aquacoelestis blue ' 
light, present perhaps the most lovely effect of colour in the 
heavens, nearly matched, however, by the variable S Cephei 
and its coerulean companion. Among numerous other ex- 
amples of contrasted or harmonising tints in double stars, 
may be mentioned : 7 Andromedse, orange and green ; 
7 Delphini, yellow and pale emerald ; ij Persei, golden and 
azure ; 24 Comae Berenices, orange and lilac ; 12 Canum Vena- 
ticorum, pale yellow and fawn; v Serpentis, sea-green and 
lilac ; a pair in Cassiopeia (2 163), copper colour and blue ; 
17 Virginis, light rose and dusky red ; Draconis, orange and 

Bright white stars have not unfrequently small blue ones 
in their vicinity. A distant companion of Regulus seems as 
if steeped in indigo ; Rigel has an azure attendant ; X Gemi- 
norum, one of an amethystine shade ; 84 Ceti and 62 Eridani 
are made up each of a white and a lilac star ; while the 
sapphire 1 Bootis is grouped closely with one, more loosely 
with two other subordinate blue objects.* 

• Flpn'Muarion, Catalogue des EtoUcs Doubles, p. 7G. 


Two questions at once suggest themselves about the 
colours of double stars. To the first, Are they real? a 
decisively affirmative answer can be given ; but the second, 
Are they permanent ? cannot be disposed of with such 
promptitude. The subtlety of hues resulting from a highly 
complex set of retinal impressions renders them peculiarly 
liable to subjective variation. As evidence of objective varia- 
tion, then, random notes of colour are of little or no use. 
Only the estimates of skilled observers, trained to the needful 
precautions, furnished with suitable instruments, above all, 
owning normal eyesight, are worth weighing and comparing. 
Under this rule of exclusiveness, the testimony requiring the 
admission of real change shrinks surprisingly in compass, 
but it does not wholly disappear. Colour-variables are to be 
found among compound, no less than among single stars. 

Owing partly to instrumental, partly to personal causes, 
the elder Struve perceived as purely white many stars seen 
by Herschel with a tinge of red or yellow. Disagreements in 
the opposite direction merited then particular attention, and 
such disagreements were especially marked in two cases.* The 
components of the splendid couple 7 Leonis were described 
by Herschel in 1784 as both white, the smaller inclining 
slightly to pale red.* But Struve saw them in 1837 golden 
yellow and * reddish ' green. Admiral Smyth ' bright orange 
and greenish yellow ; ' and such (allowing for some inevitable 
inconsistencies both in the definition and perception of colour) 
they still remain. Here then we have a strong presumption 
of genuine change, which in the companion instance of 
7 Delphini, is raised almost to certainty. These last stars, 
noted by Herschel in 1779 as both perfectly white, showed 
golden yellow and bluieh green to Struve's scrutiny. The 
progress of alteration may perhaps be marked by the younger 
Herschel's and South's record of them as white and yellowish 
in 1824.3 Their contrasted tints of orange and green now 
strike the eye at the first glance with the smallest telescope.* 

* Mensura Microm, p. Ixxxvii. 

« Phil. Trans, vol. Ixxv. p. 48. ■ lUd. vol. cxiv. p. 303. 

* Xoble, Hours with a Three-inch Telescope^ p. 111. 


-Another paii* famous for colour-fluctuations is 95 Herculis, 
composed of two equal stars of 5^ magnitude, planted (to 
appearance) immovably within 6 " of each other. Familiar 
with them as vividly tinted objects, Professor Piazzi Smyth 
was astonished, on pointing his telescope towards them fi-om 
the Peak of Teneriffe, July 29, 1856, to perceive them both 
white.^ In the following year, however, they shone as before 
in 'apple green and cherry red,' and were so observed by 
Admiral Smyth, Dawes, and others. Captain Higgens ^ 
actually watched these colours fade and revive in 1862-3, in 
the course of about a year, but no trace of them has been 
seen of late ; the stars of 95 HercuUs are of an identical 
palish yellow.' Their history goes back to 1780, when 
Herschel observed them as bluish white and white ; J. Her- 
Fchel and South called them ' bluish white and reddish ' in 
1824, Struve, 1828-32, greenish yellow and reddish yellow, in 
precise agreement with Pickering's appraisement in 1878.* 
Thus, the ' magnificent tints of orange and green ' which 
Secchi admired in 1855, and Piazzi Smyth missed m 1856, 
were of a transitory character. 

In the well-known binary, 70 Ophiuchi, there has been an 
equally undoubted change. Except an ' inclination to red ' in 
the smaller, the elder Herschel perceived no colour in either 
of these stars; his son and Sir James South called them white 
and ' livid * ; yet they were recorded by Struve as of an espe- 
cially intense yellow and purple, by Admiral Smyth as ' pale 
topaz and violet.' They are now both yellow, very much as 
they were seen by Secchi in 1855, and by Franks in 1876 ; 
the companion was, however, marked * purplish ' at Harvard 
College in 1878, * rose-coloured ' by Flammarion in 1879. 

The three stars of f Cancri are usually yellow, but Dem- 
bowski noticed them as all white 1854-6, the remoter com- 
ponent turning yellowish or olive in 1864-5.* This form 
of change is not very rare among revolving stars, both of 
which deepen and lighten, for the most part simultaneously 

* Smyth, Sidereal Chromatics^ pp. 36, 78. « Ibid. p. 80. 

* Noble, Op. cit. p. 106. * Harvard Annals, \ol. xi. p 150 
» Astr. Nach. Nos. 1110, 1574. 


through various shades of primrose and cowslip ; while the 
development in other couples of the more vivid hues of the 
spectrum tends towards the production of contrast. It 
often happens, too, that one component only varies in colour, 
in which case the change always affects the satellite star. 
The attendant for instance of B Herculis has appeared hy turns 
ashen, ' grape-red,' blue, and bluish green ; that of S Cygni 
was observed by Struve as grey in 1826-33, but conspicuously 
red in 1836, blue by Dawes in 1889-41, alternately red, 
blue, and violet by Secchi in 1856-7, grey once more by 
Dembowski in 1862-3, red by Engelmann in 1865. Since 
then it has commonly seemed light blue ; * but showed never- 
theless with the great Nice refractor in 1883 and 1886 as 
yellow or orange." Again, the multiple star <r Orionis 
includes one if not two colour-variables; the distant com- 
panion of 7 Leporis changed from pale green in 1832 to 
garnet in 1851 and 1874 ; and the satellite of v Serpentis 
from lilac in 1832 to 'native copper' in 1851.' 

M. Niesten of Brussels attempted in 1879 to assimilate 
colour- variations in revolving stars to the supposed produc- 
tion of sunspots through planetary influence.* A tendency he 
thought could be traced to the emission by both components 
of white or yellowish light near 'periastron,' or their nearest 
approach to each other, deeper or different tints becoming 
more and more prominent with their increasing distance. 
But we need only recall, in proof of the insufficiency of 
this hypothesis, that two of the most strikingly changeable 
couples — 95 Herculis and y Delphini — have 7?o perceptible 
relative movement. 

The truth is, that we are entirely ignorant as to the causes 
of the phenomena such objects present, and that the only 
kind of data by which there is much hope of their being elu- 
cidated have not yet begim to be collected. Eye-estimates of 
colour, however trustworthy, do not reach below the surface ; 
they are mere indications, which the spectroscope and the 

^ Engelmann, Asfr. Nach. No. 1G76. 

' Annales de VObservatoirc de Nice, t. ii. (Perrotin). 

=» Smyth, Sid. Chromatics, p. 29 ; Webb, Cel Objects, p. 3S9. 

* JiulL de VAcad. Bruxelles, t. xlvii. p. 50. 



' spectrograph ' (a photographic spectral apparatus) can alone 
help us to interpret. The foundation of a science of chro- 
matic variability will be laid by the first definite statement of 
a change of spectrum in connection with a change of colour. 
Until this is forthcoming, there can be no tangible ground for 
so much as a conjecture on the subject. Similarly, the 
condition sine qucL non of redness in single stars can only be 
ascertained by learning what are the prismatic alterations 
supervening upon its disappearance. We should naturally 
expect them to consist in a marked falling-o£f in the e£fects of 
absorption ; but it is also well to remember that the changes 
in colour of temporary stars have been accompanied by a 
shifting of the emissive centre of gravity , the red fading out 

Fig. 28.— Spectra of the Component Stars of fi Cygni (Hoggins). 

of the spectrum of Nova Cygni, for instance, long before the 
yellow and green. 

The spectra of permanently coloured double stars are 
marked by what we may call ' complementary absorption.' 
Dr. Huggins showed in 1864 * that in fi Cygni the yellow rays 
of the larger star escape paying toll to its enveloping vapours, 
while in its companion the blue rays are comparatively exempt 
from their encroachments (see fig. 28). A corresponding re- 
sult was obtained for a Herculis. It appears then certain that 
the contrasted colours of coupled stars are not simply an effect 
(as has sometimes been supposed) of their having reached 
different stages of growth, but arise from a profound modifi- 

» Phil Trans, vol. cliv. p. 431. 


cation of their physical state, dae in some unexplained way 
to their mutaal influence. The sky-blue, lapis lazuli, and 
yiolet tints of numerous secondary stars, unrivalled to the 
eye among solitary objects, are proved by light-analysis to 
owe their origin to exceptional atmospheric conditions. 




A DOUBLE star is one that divides into two with the help of a 
more or less powerful telescope. The e£fect is a strange, and 
might have appeared beforehand a most unlikely one. Yet it 
is of quite ordinary occurrence. Double stars are no freak 
of nature, but part of her settled plan ; or rather, they enter 
systematically into the design of the Mind which is in and 
above nature. 

The first recognised specimen of the class was ^ UrssB 
Majoris, the middle * horse * of the Plough, called by the 
Arabs ' Mizar/ which Biccioli found at Bologna, in 1660, to 
consist of a 2^ and a 4 magnitude star within fourteen 
seconds of arc of each other. Both are radiantly white, and 
they make a glorious object even in a very small telescope. 
The accident of a bright comet passing, on February 8, 1665, 
close to 7 Arietis (* Mesarthim ') led to the discovery of its 
duplex nature by Bobert Hooke in the course of his observa- 
tions on the comet. The components, each of the fourth 
magnitude, and eight seconds apart, are perfectly aUke both 
in hght and colour. Meanwhile Huygens had seen 6 Orionis 
— perceived to be quadruple in 1684— as triple in 1656; 
a Grucis, in the southern hemisphere, was divided by some 
Jesuit missionaries sent by Louis XIY. to Siam in 1685, and 
a Centauri by Bichaud at Pondicherry in 1689 ; ' making in 
all five double stars detected during the seventeenth century. 
Four more— 7 Virginis, Castor, 61 Cygni, and fi Cygni — were 
picked up in the course of the ensuing fifty-five years ; and 
in 1776, Father Christian Mayer began at Mannheim a de- 
liberate search for stellar couples. His thirty-three discoveries 

* Flammarion, L* Astronomic ^ t. ii. p. 1(54. 

M 2 


in two years might be described as the preliminary washings 
from the rich lode struck a few months later by Sir William 

The plentifulness of double stars was in itself an irre- 
sistible argument for their reality. That any two uncon- 
nected bright stars should be projected closely side by side 
upon the sphere was improbable; that such a contingency 
could be repeated blindly hundreds of times was what no 
sane man ought to have been capable of belieying. But 
human credulity is nowhere more conspicuous than in what 
it is prepared to attribute to chance ; and it needed such clear 
evidence of mutuaUy circling movements as Herschel was able 
to produce in 1808, to establish the conviction of the physiccd 
existence of double stars. 

The fact is one at which we can never cease to wonder. It 
brings us face to face with a state of things entirely unfamiliar, 
and of which the purpose lies beyond the scope of our limited 
understandings. So accustomed are we to the * sole dominion ' 
of our own great star, that the presence of two suns in one 
sphere might well at first sight appear incredible. Yet there 
are many things ' undreamt of in our phUosophy ' which are 
nevertheless true. Every drop of stagnant water is a world 
of uninterpreted mysteries ; what we choose to call the * order 
of nature ' is violated at every instant, inexplicably, by our 
own volition; and if that order be attended by anomalies 
upon the earth, how much less shall we venture to prescribe 
its course in the heavens ? 

The term ' double star ' is obviously quite indefinite, apart 
from some agreement as to its meaning ; and it was in fact 
used by early observers in a far wider sense than it is now 
usually considered to bear. Many of the small and remote 
attendants upon brighter stars recorded by the Herschels 
could scarcely be presumed to have any real connection with 
them ; 82'' was fixed by Struve as the maximum interval be- 
tween the components of a genuine double star, or 16'^ unless 
both were brighter than the ninth magnitude ; the younger 

> See the writer's History of Astronomy, p. 21, 2nd ed. 


Struve's * Pulkova Catalogue ' included no stars beyond the 
narrower limit ; and Mr. Burnham rejects all pairs below the 
eighth magnitude above 5^' apart. This progressive restriction 
has almost necessarily accompanied the improvement of 
telescopes. With the powerful and perfect refractors now in 
use, really close pairs accumulate faster than they can con- 
tinue to be observed ; and the further collection of the in- 
numerable loosely associated stellar points they reveal would 
be mere inane waste of time. 

Already some twelve thousand double stars^ in the Her- 
schelian sense, have been registered, o! which nearly half 
correspond, by the closeness of their combination, to strict 
ideas of what a double star should be; about 1,400 are 
separated by 2'' or less, and between 600 and 700 are 
visibly revolving. These last interesting cases multiply at 
present with especial rapidity. They are most apt to occur» 
as might be expected, among stars at the shortest apparent 
distances from each other, and requiring accordingly the 
highest optical powers for their detection. Our acquaintance 
with most of these is for that reason quite recent, and their 
movements are only coming to be recognised as one pair 
after another is re-measured after a few years' interval. 

The singular profusion with which stars are planted side 
by side with a bare hairbreculth of sky between, was made 
apparent by Mr. Bumham's discoveries at Chicago, while 
he still pursued the profession of a stenographer. His 
thousand new pairs included 748 at an average distance 
of l''-58.' This means that the total interval from centre to 
centre of these objects was just equal to the width of a human 
hair held thirty-six feet from the eye. About one-tenth of 
that distance is the minimum at which, even with the great 
Lick telescope, stars can be divided, but by no means the 
minimum at which they can separately exist. It is quite 
certain that numberless stars which must always, either 
through their distance from ourselves, or the closeness of 
their companions, remain opticaUy single, are nevertheless 

' Memoirs B, Astr, Soc. vol. xlvii. p. 817. 


compound ; hence of any given star, as of a chemical ' ele- 
ment/ we can say, not that it is indivisible, but only that 
it has never been divided. 

Such stellar pairs as are known to be in orbital move- 
ment are called ' binary stars/ to signify that they form real 
dual systems. The finest specimen of this kind in the 
northern heavens is Castor, or a Geminorum, composed of 
a second and a third magnitude star 5''- 18 apart. They are 
both white with a greenish tinge, and can be divided with a 
very moderate telescope, so that the sight of this brilliant 
and suggestive object is 'not reserved for the inner circle of 
astronomers. Now it happens that Bradley observed the 
relative situation of these stars in 1719, and the comparison 
of his record with measures of the present day shows that 
they have shifted in the interim to the extent of 126^, or more 
than a third of a revolution. To complete an entire one, 
then, they would need at the same rate about 420 years. 
The end of that term, however, will pretty certainly find 
them not more than half way round. And the reason is 
obvious. Throughout the last century they were travelling 
much faster than they are travelling now ; for their orbit is 
so eccentric that their greatest is twice their least distance, 
and ' periastron ' occurred in 1750. During three and a half 
centuries to come they will still continue to slacken speed, 
until at last the furthest point of mutual withdrawal is 
reached, and the slow process of approach entered upon. The 
circuit cannot be completed, it has been calculated, in much 
less than a thousand years. 

The mass of the conjoined stars of Castor could easily, 
with the help of Kepler's laws, be deduced from the elements 
of their orbit, if only their distance from the earth were known. 
But on this point we have unluckily no authentic information.' 
It seems probable, however, that they are extremely remote. 
For we can scarcely suppose the two luminaries taken together 
to contain less matter than the sun, and unless they do, they 

> Johnson^B parallax of 0"*1984 ib almost certainly erroneous. If it were 
exact, the joint mass of the stars of Castor would be less than ^ that of the 


must be at or beyond a distance from which light would spend 
forty-four years in reaching our eyes.^ But if the sun were 
thus vastly removed into space, we should receive from it 
only -j-Jir V^^ of the light we actually receive from Castor, 
which is hence likely to be a more massive and distant body 
than we have experimentally assumed it to be.^ There is no 
doubt, however, of its possessing much more than the solar 
luminosity in proportion to mass. 

The brightest is also (with one exception) the widest 
pair of revolving stars in the sky, and a third distinction — 
that of being nearer to us than any other known sidereal 
object — accounts for the first two. In a Gentauri are com* 
bined two stars so brilliant that the lesser, though emitting 
only one-third as much light as its neighbour, is still fully 
entitled to rank as of the first magnitude. Sir John Herschel 
found them both yellow, the second even brownish-yellow, 
but they are now undeniably white, and the companion-star 
must have gained considerably in lustre during the present 
century unless Feuill^e, Lacaille, Brisbane, and Dunlop erred 
egregiously together in calling it of fourth magnitude.* Since 
they were observed by the Franciscan monk, Louis Feuillee, 
at Lima in 1709, these stars have completed two revolutions 
and entered upon a third, yet there is still some uncertainty 
as to the length of their period. It seems, however, impossible 
that it should fall short of eighty-five, and unlikely that it 
should exceed eighty-eight years.^ The orbit is about as 
much elongated as that in which Faye's comet travels round 
the sun, and carries the stars, accordingly, at ' apastron ' to 
more than twice their ' periastron ' distance. They are now 
about 16^^ apart, and are separating fast, having in 1875 
swept through their point of nearest approach. The * mean 
radius,' or half the major axis of the computed ellipse, if seen 

* Corresponding to a parallax of 0"'075. 

' The mass of a star moving in an orbit of known angular dimensions grows 
with the third power, its light with the second power of the distance ascribed to it 
Mass accordingly gains upon light afl the star is pushed farther and farther 
back into space. 

* See Flammarion*s Catalogue, p. 81. 

* Powell, Monthly Notices, vol. xlv. p. 18. 


square from the earth, would subtend an angle of 18^'^ 
corresponding at the star's distance of twenty-five billion 
miles to an actual span of (in round numbers) one thousand 
miUion milesf^so that these lustrous objects are sometimes 
as close together as Jupiter is to the sun, and never as far off 
as Uranus. Their mass (computed from Doberck's elements 
including a period of eighty-eight years) is just twice, their 
light about 2^ times, that of the sun. 

The spectacle is beyond doubt an amazing one of two 
such bodies united thus organically into a single majestic 
system. That it includes many other members may be taken 
for granted, although we may never succeed in observing 
them, and are unable, even in imagination, to bestow or 
arrange them satisfactorily. Evidently, no planetary scheme 
or schemes at all resembling our own can depend upon the 
stars of a Gentauri. A Mercury or a Vulcan, at the most, 
might find shelter in the close vicinity of one from the dis- 
turbing power of the other, its possible inhabitants enjoying 
the combined or alternating radiance of a greater and a lesser 
sun. Comets entering these precincts must be perplexed to 
decide between the two potentates claiming their aUegiance, 
and perhaps on occasions pay their court to each in turns, 
throwing out tails, as they do so, in all sorts of anomalous 
and contradictory directions. It has, however, been suggested 
that the clients of double stars circulate about both simul- 
taneously, in orbits wide enough to keep them beyond the 
reach of dangerous perturbations from either. This is, of 
course, conceivable, if somewhat unlikely, nor is it impossible 
that the two kinds of scheme may be combined and harmonised 
into one highly complex system. 

The stars of 61 Gygni, like those of a Gentauri, share a 
rapid onward movement through space. They are among 
our nearest stellar neighbours, and there is nearly the same 
amount of inequaUty between them. They are, however, very 
much fainter than the southern luminaries they in some 
respects resemble, one falling somewhat short of the fifth, 
the other of the sixth magnitude. They are both, the lesser 
especially, deeply tinged with yellow. 


Although they have been under continuous scrutiny since 
17589 when Bradley noted the differences in their tunes of 
transit, it is only within the last few years that the curva- 
ture of their path has become perceptible. While marching 
alongside of it with what is called a ' common proper move- 
ment/ the smaUer also shifts its place sensibly as regards the 
larger star. But for a century and upward the shifting ap- 
peared to take place along a straight line. If this had really 
been the case, the fact would have abolished the presumption 
of their binary character, and compdled the belief, which 
was actually adopted by Captain Jacob in 1868,' that the 
stars would eventually part company and cease to have even 
an apparent connection. It is now, however, evident that 
they have not the slightest intention of doing so. The first 
symptoms of a bend in their track were noticed by Mr. Wilson 
in 1875 ; « and Dr. C. F. W. Peters's investigation in 1885 » 
not only showed them to be revolving, but set forth, at least 
provisionally, the circumstances of their revolution. 

In a period of 788 years the companion-star describes 
round its primary (supposed for the purposes of calculation 
to be at rest) an orbit of larger angular or apparent dimensions 
than those assigned to any other stellar path. Its 'mean 
radius ' measures 29^^', and this, at the distance of 61 Gygni 
(about forty billion miles), is equivalent to an interval just 
65^ times that between the earth and sun. These stars then 
circulate in an orbit more than twice as wide as that of the 
planet Neptune, and nearly as eccentric as that of the planet 
Mercury. But in the solar system a body revolving at 65^ 
times the earth's distance from the sun would finish its 
rounds in 580 years, while the components of 61 Gygni need 
for theirs close upon eight centuries. Hence their mutual 
attraction is clearly less than the power swaying the planetary 
movements. Their united mass, in fact, is less than half 
that of the sun. But since the combined surface of two 
approximately equal stars, together containing 0*45 of the 
sun's mass, amounts to 0*78 of the sun's surface, they 

» Edinburgh New PhU, Jour, vol. vii. p. 107. 

» Monthly Notices, vol. xxxv. p. 326. » Astr. Nach, Nos. 2708-9. 


ought, if equally luminous, to give 0*78 of the sun's light. 
The proportion of it, however, actually emitted by them is 
only 0-14. They are less luminous, then, per unit of surface 
(if of the same mean density) in the ratio of 14 to 78, or, say, 
five times; and, indeed, the sim in their situation would 
shine as of fully third magnitude, while they barely reach the 
fifth. These stars are hence very likely to be &r advanced in 
condensation, and may be drawing near the close of their 
career as light-givers. 

The conditions of this system would seem much more 
propitious to tranquil planetary circling than those of a Cen- 
tauri. There is more room for it, to begin with ; the stars of 
61 Cygni are six times farther apart, and their orbit is much 
more nearly circular. The perturbing influence of each upon 
the dependents of the other would accordingly be much less 
formidable than where two powerful orbs contend, as it were 
at close quarters, for gravitational pre-eminence. The planets 
of the smaller stars, too, may, on account of the mild quality 
of their radiations, nestle quite close to them without re- 
ceiving an excessive amount of heat. If the earth, for 
example, were suddenly set revolving round one of the pair 
in the Swan at the mean distance of Mercury from the sun, 
we should be conscious of no appreciable rise of temperature. 

A pair (No. 190 of Herschel and South's catalogue), 
strongly resembling 61 Cygni, is found in the constellation 
Libra. A rapid voyage in concert is in them, too, complicated 
by relative displacements which, though apparently rectilinear, 
are doubtless in reality a small section of a wide curve.' 
Comparatively little attention has, however, as yet been paid 
to them, nor are we acquainted with their distance from the 
earth. They are both yellow, and there is a disparity of 
about one magnitude between them ; hence the analogy with 
61 Cygni is physical as well as geometrical. 

Two other alUed systems are formed by the coloured pairs 
70 Ophiuchi and 97 CassiopeisB. Chromatic similarity is, 
indeed, at present impaired by the substitution of yellow for 

* Flammarion, Catalogue, p. 85 ; Burnham, Puhlica/tions Washhum Obser- 
vatory ^ vol. i. p. 130. 


the contrasting rosy hue of the companion-star in 70 Ophiuchi, 
bat it may be expected to become re-estabUshed. Both couples 
(as was pointed out by Mr. Sadler) * progress throagh space 
at about the same rate, and both are at nearly the same 
distance of twenty ' light years ' from the earth ; they show 
spectra of an identical type (the solar), and the light-power 
of each relative to mass appears to be very nearly that of 
the sun. Both, too, have proved singularly recalcitrant to 
computation. The orbit of 70 Ophiuchi, though one of the 
earliest experimented upon, can still only be regarded as 
provisionally determined. The stars have, indeed, hitherto 
so persistently refused to keep to their predicted places, that 
disturbances by invisible members of their system have some- 
times been brought in to account for their anomalous behaviour. 
The period of revolution included in the most recent set of 
elements (Mr. Gore's) is eighty- eight years ; the mass of the 
two stars is nearly three times that of the sun ; and their 
orbit is so considerably eccentric that their distance apart 
ranges from fourteen to forty-two times the radius of the 
earth's orbit. 

The still more elonp:ated path of rj Cassiopeise, traversed 
in 167 years (according to Mr. J. B. Coit's elements), is of a 
far ampler sweep. Its mean radius is 56| times that of the 
terrestrial, nearly twice that of Neptune's orbit. The stars 
are, nevertheless, at their nearest approaches, not much 
farther away from each other than Uranus from the sun, and 
since they together contain more than six times the solar 
quantity of matter, perturbations of great intensity must 
at such times affect any trains of attendant bodies they may 
separately possess. 

Laplace's conjecture that space might hold as many dark 
as bright masses has received some countenance from the 
phenomena of double stars. For among them are reckoned 
effects of the attraction of unseen upon the movements of 
seen bodies, whUe in one case the detection of an imperfectly 
luminous object has, like the discovery of Neptune, ensued 
upon the theoretical indication of its place. 

' Efiglish Mechanic t vols. xli. p. 410, xliv. p. 822. 


From the nature of the proper motion of Sirius, Bessel 
inferred in 1844 that it did not travel alone. The line traced 
out by it must, were it solitary, have been straight, whereas 
it undulated markedly and regularly once in about half a 
century. Bevolution in that period round an obscure com- 
panion was indicated ; the elements of the hypothetical Sirian 
system were computed by Peters and Auwers, and the precise 
position of the Sirian sateUite was assigned by Safford in 
September 1861. On January 81 following, it was found 
just in the right spot by Alvan Q. Clark, of Cambridgeport, 

The companion of Sinus is a dull yellow star of 8*5 
magnitude, almost lost in the glittering radiance of its great 
neighbour. Their apparent distance having diminished from 
IC in 1862 to 5" in 1889, it is now an object of such extreme 
difficulty that only a few of the best telescopes in the world 
can show it, and the only very recent observations have been 
made by Mr. Bumham with the thirty-six Lick refractor.' 
As the periastron passage is timed to occur in 1896, it will be 
some years before the pair again begins to open out. 

For many years astronomers did not feel justified in ad- 
mitting that a body so Uttle luminous as Glark*s companion 
should stUl be massive enough to sway the onward march of 
Sirius visibly to and fro. The real source of attractive power, 
it was thought, still remained to be discovered. But what 
seemed improbable has, with time, become fully established. 
The Uttle star picked up at Cambridgeport has now, during 
twenty^eight years, conformed with such general fideUty to 
the theoretical orbit of the disturbing body, as to leave no 
doubt of their identity. The system thus constituted is a 
very remarkable one. Its chief member is a body extremely 
bright in proportion to its mass ; its secondary member is a 
body abnormally massive proportionately to its light. Sirius 
shines like ten thousand, it gravitates like two of its com- 
panions. There must hence be an enormous disparity of 
temperature between them, with a probably corresponding 
difference of mean density. The smaller body may thus 

> Asir. Nach. No. 2884. 


have already advanced far on the road towards planetary 
solidity and obscurity. 

At the distance of Sirius (about fifty billion miles) the sun 
would appear as a star of the third magnitude. An accumu- 
lation of sixty-three suns would then barely supply the 
emissions of that brilliant orb, the attractive energy of which 
is, nevertheless, little more than twice that governing the 
solar system. The revolutions of its satellite are completed 
in 68^ years (adopting Mr. Gore's recent orbit), at a mean 
distance twenty-two times that of the earth from the sun, 
with excursions, at apastron, a hundred millions of miles 
beyond that of Neptune. Now to control motion so swift 
in so spacious an orbit, 8^ times the amount of matter con- 
tained in the sun must be present ; of which one-third belongs 
to the satellite itself, constituting it a body rather more 
ponderous than the sun, though giving no more than -^^ of 
its light. Thus the contrast between the components of this 
binary star could scarcely among visible objects be more 
pronounced. But a further point of strangeness is reached 
in the pairing of completely invisible with brilliantly lustrous 
bodies. This seems to be the case with Procyon, the lesser 

Since its motion is disturbed in precisely the same way as 
the motion of Sirius, there can be no doubt that it forms part 
of a binary combination. But the second member of that 
combination has not been discovered, and we may fairly say, 
Mr. Bumham's search for it in the clear air of Mount Hamilton 
with the great Lick refractor having proved fruitless, cannot 
be discovered with any optical means now available. Pro- 
fessor Otto Struve*s illusory detection of it March 19, 1873, is 
a curious example of the tricks chance now and again plays 
upon the most wary observers. The false image seen (pro- 
duced by reflections between the lenses of the equatoreal) was 
always situated about 10" from the genuine image of the 
large star, and in a horizontal line from it ; but it happened 
that the varying positions of Procyon with regard to the 
meridian at the times of the successive observations, gave 
rise, by an extraordinary series of coincidences, to just such 


an effect of revolution in the satellite as was anticipated by 
theory.* The orbit described by Procyon round the common 
centre of gravity of itself and its companion, is rather smaller 
than the orbit of Jupiter. It is nearly circular, and is traversed 
in a period of forty years. But our ignorance as to the dis- 
tance apart of the conjoined bodies entails, of course, ignorance 
as to their mass ; hence the real circumstances of the system 
still remain an enigma. Procyon is farther off from the earth 
than Sirius, though not in the proportion of its inferiority in 
magnitude. Its actual light-emissions are about one-third 
those of the greater dog-star. 

The detection of partially obscure stellar schemes opens a 
wide field to conjecture. They may be much more numerous 
than is supposed, for the undulatory movements betraying 
their existence should escape notice in any but the nearest 
stars by their minuteness, and even in these if the plane 
they lay in made any considerable angle with the line of 
sight. Here, however, the spectroscope comes to our assis- 
tance by providing the means of investigating variable motion 
in the line of sight.^ 

Nor can we avoid probing with our thoughts the relation 
obviously existing between stars like Procyon and stars like 
Algol. The common peculiarity of being attended by dark 
satellites affects motion sensibly in one case, light in the other. 
But the distinction may be more apparent than real. It is 
conceivable that the sateUite of Procyon may have originated 
in a situation resembling that now occupied by the satelUte of 
Algol, from which it was gradually pushed outward by the 
reactive effects of tidal friction. The eclipses of Procyon 
could never, however, under any circumstances, have been 
visible from the earth, for its orbit lies square to our visual 
ray, and an interposing body must needs travel edgewise 
towards and from us. 

There are other criteria besides that of visible revolution 
in an orbit, by which physical can be distinguished from 
optical double stars. Since 1812, when Bessel pointed out 

» Bull, de VAcaS, ImpMaU, t. xviii. p. 664 ; Monthly Notices, vol. xxxvii^ 
p. 193. * See infra, p. 212. 


the conclusiveness of the argument for real connection implied 
in the advance together of the stars of 61 Cygni/ ' common 
proper motion ' has been universally admitted as a proof of 
genuine association. Thus the lustrous pair 7 Arietis has 
continued relatively fixed since Bradley measured it in 1755 ; 
yet its members are fellow-travellers through space, and 
assuredly keep mutually circling as they go, although so 
slowly that a century and a quarter count almost for nothing 
in the majestic cycle of their revolutions. Again, the brightest 
star in the Southern Cross is made up of two stars of the 
second magnitude 5" apart, the situations of which have not 
perceptibly changed since Sir John Herschel determined them 
in 1884. This, however, could scarcely be the case unless both 
shared the small movement attributed to the compound object. 
And even independently of this positive test, the probabilities 
are enormously against the accidental close juxtaposition of 
two stars so brilliant and so nearly equal as those of a Crucis. 

The circumstance testifies strongly to the prevalence of 
physical connection between stellar pairs, that the average 
diflference of brightness between them grows steadily with 
their distance apart, approximately equal being thus usually 
contiguous objects. Every degree of inequality is indeed 
found in undoubted systems ; still the chances of optical 
association must obviously increase enormously, even at the 
same distances, with increase of optical disparity. 

The background of the sky is so thickly strewn with small 
stars, that we cannot be surprised if some of them happen to 
occupy critical situations. There is, indeed, more reason for 
surprise at finding that certain remote satellites of bright 
stars seem indissolubly united to them. Eegulus, for example, 
carries with it, as it pursues its way across the sphere, a star 
between the eighth and ninth magnitude, of an ultramarine 
colour, and discovered by Winlock at Washington to be itself 
closely double. The interval between the pair and its govern- 
ing body amounts to nearly three minutes of arc. Castor, too, 
has attached to it a tenth magnitude star at 74" ; and Aide- 
baran forms with a minute object at 31" what seems to be a 

' Monat. Corresixtndenz, JBd. xxvi. p. 160. 


permanent combination resembling in its effect to the eye 
that of Mars with his outer satellite.^ 

Where two close stars seem fixed, relatively and absolutely, 
the case for their physical union must depend upon circum- 
stantial evidence alone. But this is sometimes of overwhelm- 
ing strength. Contrast of colour, for instance, may afford 
grounds for a persuasion amounting almost to a certainty 
of real relationship. Such tints as blue, green, and violet, 
only occur among mutually associated stars; nor can we 
possibly suppose the association upon which they depend for 
their production to be merely apparent. The topaz and azure 
components of /9 Cygni, for instance, have no appreciable 
motion of any kind, and they are separated by a gap of 34^', 
exceeding the limit of distance of real double stars as defined 
by Struve. But it is impossible to doubt that their brilliant 
hues are truly expressive of the systemic union from which 
they in some unknown way result. The same may be said 
of B Gephei and its bright blue attendant, and of the much 
closer and nearly equal stars 95 Herculis, the inference in 
their case being strengthened by the concerted changes of 
colour recorded of them. We might be sure, too, of the 
dependent status of the emerald satellites of the red stars 
a Herculis and Antares, even if it were not independently 
proved by a community with their primaries in very slow 
progressive movement. 

Nevertheless, some highly coloured pairs have been con- 
cluded by M. Flammarion, from a careful study of their 
relative displacements, to be optical.' Among these are 
o Draconis, which might be called a replica, subdued in lustre, 
of fi Cygni ; o-* Piscium, golden and blue ; 42 Piscium, 
yellow and green ; a gold and purple pair in Auriga (02 154) ; 
two nearly equal stars in Perseus (2 434) showing lovely tints 
of golden yellow and azure, with several besides. 

If these colours be inherent, it is difficult to believe that 
the stars distinguished by them are simply thrown together 
by perspective. Before venturing to pronounce, however, we 

1 Burnham, Astr. Nach. Nob. 2189, 2875. 
* Comptes RendiM, t. Ixxxvii. pp. 836, 872. 


must wait, and let their motions develop. By rashly ant:- 
cipflting nature, we often only display our ignorance of her 
hidden workings. Meanwhile, experiments might be made 
with a view to detect any lurking influence of contrast upon 
the hues of these particular couples, which, however regarded, 
present points of curious interest. 

The display of similar goes quite as far as the display of 
dissimilar colours of an unusual kind, towards proving a 
physical union between adjacent stars. Strikingly red pairs, 
for instance, even when pretty widely separated, can hardly 
be the result of chance. Several of them are known, but 
their fixed character does not invite frequent observation. 
Variability of light supplies another valuable indication of 
relationship. When common to both members of a pair, it 
kaves no room for doubt, on the subject. We shall recur to 
this topic in the next chapter. 

Stars with ascertained proper motions characterise of 
themselves the nature of their companionship. For either 
they keep on together, or they show signs of incipient separa- 
tion, and so slowly but surely discriminate between a lasting 
union and mere temporary contiguity. In the latter case 
the movement of one of the stars referred to the other neces- 
sarily proceeds along a straight line, so that rectilinear dis- 
placement is an infallible, and the only mark of an optical 
couple. One curiously close (2 1516) occurs in the constella- 
tion Draco. Two stars, of 7 and 7*5 magnitude, passed in 
1856 so near to one another by the hazard of their paths 
nearly intersecting, as to present the effect of two points of 
light, one inch apart, at a furlong's distance from the eye. 
Their angular separation, then only 2"-6, is now 13'^ and it 
will continue to grow indefinitely. Their absolute disconnec- 
tion has been confirmed by direct measurements showing them 
to difi'er enormously in remoteness from the earth. The 
larger of the two, by one of those singularities which abound 
in the heavens, forms a genuine pair with a star very much 
fainter than its spurious companion. 

From what has been said, it is clear that a good deal of 
patience is needed for the investigation of double stars. The 



facts about them must often be allowed to ripen for a long 
time before they can be turned to account. Sooner or later, 
however, their fruit cannot fail to appear. There is, perhaps, 
no other branch of science in which industry is so sure to be 
rewarded with definite results. The first stop is to separate 
clearly perspective from physical couples ; and this can only 
be done by the persistent repetition of exact measures. The 
next is to detect circulatory movements in the latter as they 
begin to be aj)parent, or to keep watch on them as they progress. 
Their careful comparison with theory may at any time bring 
surprising novelties to light. For each stellar system is in 
effect a world by itself, original in its design, varied in its rela- 
tionships, teeming with details of high significance. But at 
present only an imperfectly traced outUne of the construction 
of some three score among thousands of them, is before us ; 
their multitude, in fact, distracting attention. Yet it would be 
better to make intimate acquaintance with one than to know 
hundreds superficially. All the resources of modern inven- 
tiveness should be enUsted in these inquiries. Not only the 
revolutions, distances, and masses of double stars, their move- 
ments across and in the line of sight, should be determined 
with ever-increasing precision, but their colours and magni- 
tudes, and above all, their separate spectra, both visual and 
photogi'aphic, should be recorded. By such means as these, 
real knowledge will be augmented far more than by the most 
brilliant success in the telescopic detection of new pairs. 
This has its own interest and value, but the recesses of 
sidereal structure must be otherwise explored. 




The light-changes of double stars are commonly of a fitful 
and indecisive kind. They may affect one or both members 
of stationary pairs; but visibly revolving stars, as a rule, 
conspire to vary, if they vary at all. The alternating fluc- 
tuations of 7 Virginis, discoverable only by close attention 
to the swaying balance of lustre between the components, are 
in this respect typical. Each may be described as normally 
of the third magnitude, and each in turn declines by about 
half a magnitude and recovers within a few days, yet so that 
the general preponderance during a cycle of several years, re- 
mains to the same star. The existence of this double periodi- 
city was recognised in 1861 by M. Otto Struve, who, how- 
ever, despaired of investigating it with success in a latitude 
where the stars in question never rise more than 80° above 
the horizon.' 

Their circulation is in the most eccentric of ascertained 
stellar orbits (see fig. 30). The ellipse traversed by 7 Virginis 
in 180 years ia, in fact, proportionately somewhat narrower 
than the path round the sun of Encke's comet, so that the 
stars will in 1926 be separated by fully seventeen times the 
interval of space between them in 1836, when they merged 
into a single telescopic object. Their inequaUties of light 
seem to have developed as they approached each other; at 
least, they first began to be noticed by Struve in 1818, and 
they at present tend to become obliterated, whether to revive 
with regained proximity towards the close of the twentieth 
dcntury, future observations must decide. A spectrum of the 

' Observations dc Potilkouia^ t. ix. p. 122. 

.N 2 


Sirian pattern is combined with a perceptible tinge of yellow 
in their light. 

Belative variability is in 44 Bootis still more marked than 
in 7 Virginis. But here a fundamental disparity between the 
components is seldom and temporarily abolished. Noted by 
Herschel as considerably unequal in 1781, they appeared to 
him perfectly matched in 1787. And it may be noted that 
they had in the interim passed periastron. Btruve observed, 
June 16, 1819, a difference between them of two magnitudes, 
which had sunk to half a magnitude in 1833. Argelander 
found them precisely equal June 6, 1830 ; Dawes perceived, 
April 27, 1841, a slight advantage on the side of the usually 
smaller star ; while the superiority of its companion was recorded 
by M. Duner at Lund as ranging, during the years 1869 to 
1875, from 0*4 to 1'3 magnitudes.* Since their changes are 
often simultaneous, though not always in the same direction, 
their combined variability has never been conspicuous. The 
stars of 44 Bootis complete their rounds in a highly eccentric 
orbit in a period of 261 years.* Their tints, varj'ing from 
yellow and sky-blue to white and dull grey, cannot be without 
influence upon their photographic magnitudes, which were 
determined at Paris in 1886 to be 5*3 and 6. Their joint 
light, though of the same spectroscopic quality, has then 
only one-twelfth the intensity of that of 7 Virginis. 

The component stars of f Bootis when photometrically 
measured at Harvard College in 1883 were of 4*4 and 4*8 
magnitudes, but the order of their brightness has been at 
least three times reversed during a century of observation.' 
Their period of revolution must be of enormous length. 
From 1796 to 1841 they appeared fixed ; then a very slow 
wheeling movement became perceptible, accompanied by a 
diminution of distance, and it now taxes the powers of the 
best telescopes to divide them.^ Their spectrum is of the 
Sirian type. 

' Lund Observations^ 1876, p. 74. . « Doberck's elements. 

• JIaiTard Annals^ vol. xiv. p. 458 ; Observations de Foulkowa, t. ix. p. 
143 ; Dun^r, M^surcs MicronUlriqueSy p. C8. 

* Crossley, Handbook of Double StarSy p. 299; Tarrant, Jour Liverpcol 
Astr. Society, vol. v. p. 77. 


An analogous object is a Piscium, made up of a fourth and 
a fifth magnitude star at 3" distance, and revolving in a period 
unUkely to be much less than two thousand years. The 
larger certainly varies in light, and perhaps also in colour, 
the smaller certainly in colour, and perhaps also in light.* 

An observation made by Mr. Tebbutt in New South Wales, 
August 22, 1887, gave a unique proof of the relative varia- 
bility of a close double star in Virgo (OS 256). At its occulta- 
tion by the moon on that night, the chief part of the light 
went out with the disappearance of the reputed lesser star, 
the component which had of late passed for its primary 
remaining still for a few moments separately but dimly 
visible.'* Similar but less marked reversals had already been 
noticed by 0. Struve and Dembowski in this slowly circulating 

Variability, as we have said, affects both or neither of two 
stars so intimately united that their orbital movements have 
become apparent after a comparatively short lapse of time, 
A possible exception, however, to this rule is met with in 
B Cygni. This beautiful and delicate pair was discovered by 
Sir William Herschel in 1783, but in 1802 and 1804 he 
totally failed, with improved optical means, to see the eighth- 
magnitude companion. His son was equally unsuccessful 
under the best atmospheric conditions in 1823, and Sir 
James South and Gambart in 1825.* It emerged to view, 
however, with Struve's nine-inch Fraunhofer in 1826, and 
has since been rarely missed. An occultation of one star by 
the other, postulated to account for the telescopic singleness 
of the pair between 1802 and 1826, was by their subsequent 
movements decisively shown not to have taken place, and 
the alternative hypothesis of a temporary loss of light in the 
small component was, almost of necessity, adopted. Yet it 
has received no strong countenance from recent observations. 
M. Duner acquired the conviction from seven years of expe- 
rience that the visibility of an object at all times diflBcult 

* Harvard Annals t vols. xi.p. 112, xiv. p. 433 ; Flammarion, CataJogtiP, p. 12 

* Observatory, vol. x. p. 391. ■ Obs. de Poulkowa, t. ix. p. 327. 

* PhiL Trans, vols. cxiv. p. 339 ; cxvi. p. 376. 


depends entirely upon the state of the air ; ^ and Mr. Burnham 
seems to be of the same opinion.* 

Changes of colour in this satellite star are nevertheless 
patent. Struve found it of an ashen shade from 1826 to 1883 ; 
in 1836 of a bright red.' It has since generally appeared blue ; 
but Duner saw it once olive, though otherwise always red ; 
and intervals of greyness are also on record. The computed 
orbits have hitherto failed to represent the movements of the 
system with any degree of accuracy ; but Mr. Gore's with a 
period of 377 years may prove more successful. 

Relative variability has recently been detected by M. 
Flammarion in 7 Arietis ; it is, or has been, also present in 
6 Serpentis, 38 Geminorum, tt Bootis, e Arietis, and many 
other couples, most if not all of which give spectra of the 
Sirian type. Their agreement in the possession of this par- 
ticular quality of light is the more remarkable from its being 
the badge in solitary stars of exceptional emissive stability. 
Every 'white star,* so far known to be variable, has proved 
also to be compound, and those of the Algol type are so far 
from making an exception to this rule, that they are among 
the most rapid of possibly existing binaries. 

Besides these, we are acquainted with only two Sirian stars, 
h Orionis and S Monocerotis, which have had periods of light- 
change assigned to them. The first is a wide double star of 
dubious variability ; the second is the leading member of a 
straggling cluster, and was discovered by Winnecke in 1867 
to change from 4-9 to 5*4 magnitude in 8d. lOh. 88m. Of 
two close attendants^ the inner and brighter, at a distance of 
2"-8, has been thought to be in slow circulation, but the point 
is still unsettled* The system has no appreciable proper 

A star situated near a Virginis is the only object charac- 
terised by a spectrum of the first type known to undergo 
extensive intrinsic fluctuations of light. From a comparison 
of observations going back to the tenth century, when Al Siifi 

* Mdsures Micromitriques^ p. 119. 

' Westwood Oliver's Astronomy for Amateurs^ chap, vii 

' * Color comitis egregius/ Mens, Microm, p. 297. 


Ifegistered it as of fifth to sixth magnitude, Schmidt ascertained, 
in 1866, its irregular variability from the fifth to the eighth.^ 
The anomaly of such changes in a Sirian star was brought 
more into harmony with other examples by Burnham's 
division of ' Y Virginis,' at Chicago in 1879, into two nearly 
equal components, less than half a second (0'''47) apart.^ A 
subsequent observation gave no satisfactory evidence of altera- 
tion, either in brightness or position, during the intervening two 
years.^ This is not surprising, since accesses of light-change 
in double stars are very generally followed by long intermis- 
sions. Nor could orbital motion, although presumably in 
progress, be expected to become so quickly aj)parent. The 
study of its laws, and of the varying magnitudes of the 
members of this singularly interesting system, ought not to 
be neglected by the possessors of great telescopes. 

The fluctuations of U Ta.uri (observed by the late Mr. 
Baxendell, 1865 to 1871), Uke those of Y Virginis, seem for the 
present suspended. The twofold nature of this object, which 
is situated quite close to a variable nebula, was detected by 
Mr. Knott, December 4, 1867. Each component is of 9*7 
magnitude, and they lie 4^'' asunder.* No recent observations 
of them that we are aware of have been made. 

The variotinted pairs 8 Cephei and ^ Cygni both belong 
to the class treated of in the present chapter. The former is 
the well-known short-period variable with which we have 
earUer become acquainted ; * the latter changes slowly and 
almost imperceptibly between 3*3 and 3-9 magnitude.^ The 
satellite is, in each case, exempt from the suspicion of insta- 
bility. Not so the fifth magnitude attendant of a Herculis. 
The elder Struve considered that it declined occasionally to 
one-sixth its normal brightness;^ and Father Secchi also 
perceived irregularities,® which have, however, for many years 
past ceased to be noticeable. The green hue of this star 

* Astr. Nach, No. 1697 ; Harvard AnnaU^ vol. xiv. p. 456. 
' Observatory^ vol. iii. p. 192. 

• Mems. R, Astr. Society^ vol. xlvii. p. 190. * Ibid. vol. xliii. p. 78. ' 
» See ante, p. 132. • Klein, Astr, Nach. No. 1663. 

' Mtnsurce Micr(ymetric:Sy p. 97. * Atti dclVAccad. Pont, t, vii. p. 62. 


(although to many eyes appearing blue) is in reality quite 

The variations from 6 to 6*8 magnitude of U Puppis, 
detected by Mr. Espin in 1883/ derive added interest from 
the strong probability that they integrate the changes of two 
close components. The star is the chief member of one of 
Struve's wide fixed pairs (2 1097) ; after being ' elongated * by 
Dembowski in 1865, it was fully resolved by Bumham in 
1875 into two unequal objects at an interval of 0'''80. There 
is no other short-period variable of so obviously compound a 
nature. Its spectrum is of the solar type. 

A corresponding long-period star is 17 Geminorum, per- 
ceived by Mr. Burnham, during a visit to Mount Hamilton in 
1881, to form 'a splendid unequal pair, likely to prove an 
interesting system.* * Its revolutions will deserve the more 
attention that no star showing a banded spectrum has yet 
given perceptible signs of orbital movement. As a variable, 
71 Geminorum may be described as an abortive specimen of 
the Mira class. Its phases, run through in a period of 229 
days, are always ill-marked, and at times almost wholly sus- 
pended. The share of the companion in bringing about these 
partial effacements has yet to be determined. Its changes, 
for instance, may possibly be to some extent of a compensa- 
tory character; or its influence may tend to interrupt the 
regular progress of those of its primary. The latter seems the 
more likely alternative. 

The presumption of sympathy in variability between closely- 
conjoined stars, although supported by many facts, has not 
yet been tested with any degree of strictness ; but the con- 
verse proposition, that agreement in light-change implies 
physical connection, is of all but self-evident truth. Two 
variables in Cygnus,^ for example, situated 24" apart, may 
safely be assumed to constitute a system, their * ruddy and 
coerulean ' tints being a confirmatory circumstance. A still 
more striking combination is presented by U Cassiopeife and 

* Monthly NoilceSt vol. xlvii. p. 432. 

' Ibid. vol. xlvii. p. 204 Burnhara's measures at Lick, in 1889, sugges 
orbital movement, see Astr. Nich. No. 2930. * /( 1470 »Lalande 3B42». 


a blue companion at 59'', with which its strong red glow 
contrasts at times splendidly. The principal star fluctuates 
irregularly from the sixth to below the ninth magnitude ; the 
attendant from the eighth to the tenth. The probability of 
their being united by a special tie is overwhelming. Accordant 
variability of a conspicuous kind is an argument for its 
existence to the full as convincing as the possession of a 
common proper motion. 

The crimson tint of U Cygni, discovered by Mr. Knott in 
1871 to vary from above the eighth to below the eleventh 
magnitude in a period of 466 days, was described by Webb as 
' one of the loveliest in the sky.' It is set off by the blue 
rays of a companion at 63", which seems to fluctuate in 
colour, but little, if at all, in light. Their azure is, however, 
no mere, optical effect of contrast, since (though capable of 
fading independently)* it survives without alteration the 
telescopic extinction of the adjacent red luminary. U Cygni 
is the only star belonging to the fourth spectral order open to 
a suspicion of being in systemic connection with a neighbour. 

A good many variables have satellites as to which no such 
suspicion arises. Thus, the ninth magnitude star within 46^' 
of the beautiful * carmine '-tinted object S Orionis is undis- 
tinguished either by colour or change, and they hence very 
likely form only a perspective couple. The same inference 
applies to tnree small stars contiguous respectively to 
E Cassiopeia, E Crateris, and Mira Ceti, detachment through 
the proper motion of the variable being, in the last case, 
visibly in progress. 

The light-changes of connected stars are of great import- 
ance to the theory of stellar variability. For since they affect 
objects which the character of their spectra warrants us in 
saying would shine steadily if single, their mutual influence 
is manifestly concerned in generating fluctuations of a certain 
ill-pronounced type. This relation is rendered the more 
significant through the possibility, brought into view by 
Mr. Lockyer's meteoric theory, that variability of every 

* Birmingham, Trans. B, Irish Acad, vol. xxvi. p. 300 ; Gemmill, English 
Mechanic^ vol. xlyi. p 340. 


kind may depend for its production upon external action by 
closely-circulating and to us invisible bodies. A test too may 
be furnished by fluctuating couples to the opinion that 
luminous instability belongs to a late stage of stellar ex- 
istence. The contemporaneous origin and similar constitution 
of members of binary systems are indicated to our minds as 
highly probable. If this be so, development should proceed, 
other things being equal, quickest in the smallest masses, more 
slowly in the larger.' Hence, if it were true that variabiKty 
accompanied decline, companion-stars should be far more 
unstable in light than their primaries. Facts, however, 
contradict this inference. Mr. Burnham is altogether in- 
credulous as to the alleged disappearances of certain satellites 
through loss of light ; and in every undoubted case of varia- 
bility in one member only of a pair, the member it distinguishes 
is the principal star.^ This circumstance gives a pregnant 
hint on the obscure but eminently interesting subject of the 
origin, history, and mutual relations of conjoined stars. 

* Lockjer, Proc, B, Society y vol. xliv. p. 90. 

' For a list of variable Double Stars, see Appendix, Table II. 




The strong presumption that the law of gravitation would 
prove truly universal has been fully borne out by investiga- 
tions of stellar orbits. Binary stars circulate, it can be un- 
hesitatingly asserted, under the influence of the identical force 
by which the sun sways the movements of the planets, the 
earth the movements of the moon. It is true that this does 
not admit of mathematical demonstration, but the over- 
whelming iraprobabihty of any other supposition amounts 
practically to the same thing.* The revolutions of the stars 
are hence calculable, because conducted on familiar principles ; 
their velocities have the same relation to mass, their perturba- 
tions may lead to similar inferences as in the solar system. 

Observations, however, must precede calculations; and 
they are rendered arduous in double stars by the extreme 
minuteness of the intervals to be measured. Many revolving 
pairs never separate to the apparent extent of a single second 
of arc ; yet this fractipn of a second may represent, in abridg- 
ment, a span of some thousands of milhons of miles. Infini- 
tesimal errors, magnified in this proportion, become of 
enormous importance, and often impenetrably disguise the 
real aspect of the facts. 

For determining the relative situations of adjacent stars, 
two kinds of measurement are evidently needed. The first 
gives their distance apart, the second the direction of the line 
joining them as regards some fixed line of reference. That 
selected is the * hour-circle,' or great circle passing through 
the pole and the larger star, and the angle made with it by 
the line of junction of the pair is called their * position-angle/ 

' Tisserand, BiUletin Astronomique, t. iv. p. 13. 

188 THE system: of the staks 

It is counted from 0° to 360°, in a direction opposite to that of 
the movement of watch-hands ; and a star is hence said to be 
in direct revolution if its circuit is from north to south 
through east, but in retrograde revolution, if it is oppositely 

Now the successive places, from year to year, of the moving 
star, obtained in this way with absolute accuracy, would 
fall into a perfect ellipse, the foreshortened delineation of 
the real ellipse traversed in space. For the star is seen 
by us projected against the sky, or rather upon the plane 
touching the sphere at that point, while the actual orbit may 
lie in any one of an infinite number of planes. The two 
curves, nevertheless, have relations by which one can be 
deduced from the other with geometrical certainty. Both are 
ellipses, and in both the * radius vector,' or line drawn from 
the satellite to the primary, describes equal areas in equal 
times. But the position of the chief star at the focus of the 
real ellipse is not maintained in its perspective representation, 
in which the * projected focus ' is often quite unsymmetrically 
placed. Once then the seeming orbit of a binary star is 
thoroughly ascertained, the problem of determining its actiial 
orbit is as good as solved, the transition from one to the 
other being effected by a mathematical operation of no 
considerable difficulty. Even when the seeming orbit is 
a straight line the process remains feasible, and in fact one 
of the most reliable stellar theories relates to a couple, the 
movements of which are conducted in a plane passing, it may 
be said, accurately through the earth.* The stars of 42 Comae 
Berenices appear simply to oscillate to and fro in a period of 
somewhat less than twenty-six years, never diverging to a 
greater extent than about half a second, and occulting each 
other completely twice in the course of a revolution. Dis- 
covered by Struve in 1827, they have since five times pre- 
sented an aspect of indivisible singleness. Other mutually 
occulting pairs are y Coronse Borealis with a period of ninety- 
five, 44 Bootis with one of 261 years, and a binary in Ophiu- 

> 0. struve, Monthly Notices, vol. xxxv. p. 367 ; Atti delV Accad. Pont, t. 
xix. p. 259. 


cbuB (2 2178) of somewhat fluctuating brightness, revolving 
in forty-five years. 

Nothing would at first sight seem easier than to lay down, 
from a sufficient store of data, the apparent path of a star. 
Yet the task is often a most embarrassing one. Owing to 
the excessive minuteness of the quantities concerned, the 
best observations can give only loose approximations to the 
actual facts. The margin of uncertainty, always very wide, 
at times exceeds any reasonable limit, and computers are 
hence obliged, as a rule, to reject some part of the materials 
before them as misleading and incompatible with the rest. 
But the exercise of discretion leads to diversity of resi^lts, and 
totally different orbits can thus be derived from the same set 
of observations, by varying their treatment so as to distribute 
differently their inherent errors. Where only a moderate arc 
of the orbit has been seen to be described, the problem of 
ideally completing it admits, from the indeterminateness of 
its conditions, of no rigid solution. When the companion of 
Sirius, for instance, had been eighteen years under scrutiny, 
it was still impossible to decide whether it would return to its 
starting-point within the half century allotted to it on grounds 
independent of its* visible movements, or depart on a remote 
excursion from its primary, demanding some hundreds of years 
for its accomplishment.* 

In no department of astronomy is the. mischief of 'per- 
sonal equation ' so sensible as in the measurement of double 
stars. Nearly all available data are prejudicially affected by 
it, and those emanating from different individual sources are 
thus often rendered exceedingly inharmonious. Much labour 
and ingenuity have been spent in determining its direction 
and amount for various observers, with a view to freeing their 
results from its effects ; and after all, it remains a question 
whether the observations so elaborately corrected are not more 
misleading than in their * raw ' state. 

All these complications can be at once swept away by 
substituting the camera for the micrometer. The photo- 
graphic method leaves no room for systematic, very little for 

* Pluminer, Monthly Notices^ vol. xlii. p. 63. 


accidental errors. G. P. Bond, of Cambridge (U, S.), showed 
in 1857, long before the introduction of the modern ' dry 
plates,' its wonderful capabilities for the accurate registration 
of the varying relative situations of double stars ; * and those 
capabilities were more fully realised in 1886 by the skill of 
the MM. Henry. We are permitted to give in iSg. 29 some 
specimens of the Paris photographs of double stars. The re- 
peated impressions shown of each pair were obtained by 
allowing free play to the diurnal motion during certain de- 
finite intervals between successive short exposures. Thus the 
line of displacement of the stars traces out part of a circle of 

K Bolitifl . ^^^^^^^^^^^^■^^■^^■^^■^^■^^■^^1 13" 9 

68 . ^^^^^^^^^^^^^^^^^^^^^^^^^^^^H 5" 7 

y Vir^nis ^^^^^^^^^^^^^^^^^^^^^^^^^^^M »"34 

Fio. 29.— Four Doable Stars photographed at Paris. (From Moachez's 
* Photographic Astronomique.') 

declination, and their angles of position are directly measurable 
from plates, embodying the data for their own orientation. 
The exactitude of determinations from them proved very 
remarkable ; for ? UrssB Majoris the * mean error ' of single 
measures of distance amounts to only 0"-077, of angle to 
0'°55.^ And this is no illusory precision, undermined by 
evasive uncertainties, but the statement of a fact hard 
enough not to crumble in the handling. Stellar orbits will 
then really be known when they come to be calculated solely 
from photographic m3asures. 

> Astr. Nach. No. 1129. 

« Mouchez, La Photographic Abironomiquc, p. 44. 


Their application is, however, at presant restricted to such 
pairs as are neither very unequal nor very close. The 
diffusive brightness of Sirius, for instance, leaves no possi- 
bihty of getting a separate print of its companion ; nor could 
even the much lesser disparity between the stars of S Cygni 
be m^de compatible with the distinct self-portraiture of both. 
Again, the minimum interval at which even perfectly equal 
couples have hitherto been successfully photographed exceeds 
two seconds, closer objects running together on the plate. As 
experience and invention progress, these limitations may be 
removed, but they are as yet eflfectual. 

The systematic adoption of the new method promises to 
bring material reinforcement to the resources of the science 
of compound stars. Observation will be enabled by it to keep 
pace with discovery, and stars need no longer be found only 
to be forgotten. The exposure of a sensitive plate during a 
few seconds yearly to the rays of each couple will supply, 
after a time, a stock of facts impaired in value by no per- 
plexing inconsistencies. The apparent orbits of revolving 
stars will be virtually inscribed, for the benefit of computers, 
in each tolerably complete collection of negatives. In ' smooth- 
ing the curve,' a process always inspiring distrust, little of 
arbitrary choice will. remain ; and the representative ellipse, 
instead of threading its way amid a straggling crowd of out- 
lying observations, will pass, not, indeed, actually through 
(which would imply the annihilation of error), but very close 
to all the given places. 

About sixty stellar orbits have so far been computed by 
Dr. Doberck, Mr. Gore, MM. Glasenapp, Duner, Celoria, and 
some others. But for the most part tentatively only, nor 
always with success. Predictions are often very far astray, the 
moving stars showing themselves totally regardless of the 
trammels of theory. Thus, the equal members of the lucid 
pair 7 Coronie Australis had, in 1887, run ahead of anticipa- 
tion to the extent of twelve degrees ; and the period of 
6 Eridani, another fine southern binary, has had, owing to 
grave discrepancies with the Sydney observations of 1871 to 
1881, to b3 lengthened from 117, first to 224, then to 302 


years. Nor can we feel much confidence that the path of 61 
Cygni is really that traced out by Peters, or that Castor will 
duly complete the millennial course prescribed to it by Thiele 
and Doberck. But, as we have said before, the cause of these 
uncertainties is to be found in the limits placed upon visual 
accuracy by the conditions of our existence, not at all in any 
want of ability on the part of the computers. 

Satisfactory acquaintance has hardly yet been made with 
the movements of more than half-a-dozen stellar pairs. Those 
of the quicker kind naturally exhibit their nature soonest, 
and indeed revolutions require to be finished, or nearly 
finished, before they can be said to be ascertained. Among 
the best stellar theories extant is that of f UrsaB Majoris, one 
of two fourth magnitude stars marking the hindmost paw of 
the Great Bear. Divided by Herschel in 1780, this couple 
was made by Savary, in 1828,' the subject of the first experi- 
ment in the extension of Newtonian principles to the sidereal 
universe. It succeeded ; for the stars were found to describe, 
as nearly as could be expected, the orbit calculated for them 
on the supposition that their mutual gravitation was the 
influence binding them together into a moving system ; and 
the validity of Newton's law wherever matter exists has 
never since been open to serious question. 

The path of f UrscB has of late been several times re- 
investigated, and with results so concordant as to give a 
strong assurance of their approximate accuracy. It is a 
considerably elongated ellipse, the eccentricity being expressed 
by the fraction |, which is just twice that of the orbit of 
Mercury, half that of the orbit of Encke's comet. The period 
of traversing it is 60^ years ; its semi-major axis would 
subtend, if seen without foreshortening, an angle of 2^", and 
it lies in a plane inclined 66° or 57° to what we may call the 
ground-plane of the heavens— the tangent-plane, that is, to 
the sphere at that point.' 

We are ignorant of the mass of this system because we 

» Conn, dcs Temps, 1830, pp. 56, 163. 

* Data on these aeveral heads, together with others defining the situation of 
periastron, and of the line of interp^rtion of the orbital and reference planes, 
constitute what are called the • elements ' of a star's movements. 


are ignorant of its distance from the earth ; but if we were to 
assume, by way of illustration, that it is at what Struve 
somewhat precariously estimated to be the * average distance * 
of a fourth magnitude star,* it would follow, from the vast- 
scale of its construction combined with the rapidity of its 
movements, that the gravitalional power residing in it exceeds 
194 times that of the sun. There is, indeed, a strong Ukeli- 
hood that these figures exaggerate both the remoteness and 
the massiveness of f Ursee, but whatever its distance and 
whatever its mass, we can say with certainty that it is an 
intensely luminous body. If of the same mean density, it 
must be, square mile for square mile of surface, of about two 
and a half times the solar brightness. 

Determinations of the distances of binary stars are of special 
interest from their leading to a knowledge of their masses. 
The connection is easily explained. Angular measurements, 
which are the only ones possible to be got of objects out of 
tangible reach, are convertible into definite linear values when 
the radius of the sphere they refer to becomes known — in 
other words, when the interval of space between the eye and 
the objects measured is ascertained. So that the dimensions, 
in seconds of arc, of the orbits of stars at measured distances 
give at once their dimensions in millions of miles, whence, 
with the help of the periods of the revolving objects, their 
masses easily follow. For by the law of gravity, the attractive 
power of any system is proportionate to the cube of the mean 
distance of the bodies composing it, divided by the square of 
their period. Employing, then, as a unit of space in this 
little calculation, the distance of the earth from the sun, and 
the year as our unit of time, we get the mass of each pair of 
revolving stars in terms of the sun's mass. It comes out, of 
course, larger in the ratio of the cube of the distance for the 
same period, and smaller in the ratio of the square of the 
period for the same distance. Swiftly-moving and spacious 
systems contain accordingly great quantities of matter; 
sluggish ones comparatively little. 

* EtaUes Doubles^ p. 46. From Madler's elements, Struve deduced a value 
for the mass of 4 UrsaB equal to 159 times the solar mass. 


The quickest of known binaries is S Equulci, divided at 
Pulkowa in 1852 into two stars of about 4^ magnitude, set 
so close together that only the very best instruments can show 
them separately. Nor can even they do so at all times. The 
stars move in a plane almost coincident with the visual ray, 
and seem consequently to perform little vibrations, carrying 
them apart, at intervals of approximately seven years, to the 
extent of 0"'44. During some intervening years they are 
optically merged into one. The difficulty of rightly inter- 
preting such inconspicuous appearances is obviously very 
great, and has not yet been overcome, recent measurements 
by Mr. Bumham showing Wrublewsky's period of 11^ years * 
to be decidedly too short.^ The true period of 8 Equulei 
will probably be found not to differ much from fourteen years. 

The pair most nearly approaching its rapidity is y3 Del- 
phini, discovered by Burnham in 1873, and completing its 
revolutions, by the recent investigations of M. Celoria, in 
seventeen years ; while ? Sagittarii, first divided by Professor 
Winlock in 1867, comes third with a period of 18^ years. 
The * occulting stars * in Coma Berenices (Flamsteed's No. 42) 
circulate in 25^, ^ Herculis and an interesting couple in Leo 
(2 8121) each in a few months less than thirty-five years. 
Among stars at known distances from the earth, 85 Pegasi 
has the shortest period (twenty-two years), and its mass 
proves to be eleven times that of the sun. These stars, as 
Mr. Gore remarks, present rather the appearance of a sun 
and planet than of two suns. The primary centuples the 
light emitted by its satellite, and there is just the disparity 
between them that would be presented by the sun and Jupiter 
if of the same intrinsic brilliancy. These would, on the con- 
dition supposed, constitute, at the distance of sixty light-years 
(attributed to 85 Pegasi), a pair of respectively 7i and 12^ 
magnitude, never above 0'''28 asunder. The utmost powers 
of the great Lick refractor would, however, scarcely be adequate 
for their separation. 

In all, eleven star-couples have had periods assigned to 
them of less than fifty years, fifteen of less than a hundred 
but more than fifty. The slowest of computed binaries is 

* Glasenapp, Astr. Nach, No. 2771. * Ibid. No. 2875. 


^ Aquarii^ a bright, nearly equal pair just in the equator, 
needing, according to Dr. Doberck, 1578 years to finish its 
circuit. But there is no certainty on this point. Half a dozen 
totally diBferent orbits could probably be accommodated to 
the arc of 45° described since 1779, equally well with that 
provisionally fixed upon. 

But even these leisurely movements are swift compared 
with others which, after the lapse of upwards of a century, 
seem barely nascent or even non-existent. The apparent 
fixity, indeed, of stars at the considerable intervals separating 
the components of fi Cygni, S Cephei, Serpentis, and others, 
accords with what we know of the prodigious scale of sidereal 
construction, but the indication for moderately close pairs 
of periods ranging up to or beyond twenty thousand years is 
startling. Such are 95 Herculis, a in the same constellation, 
and 7 Arietis ; the movements of ? Ursie Majoris, causing a 
change of angle of five degrees in 135 years, suggest their com- 
pletion in about 10,000, and 4,000 may sulBfice for those of 
7 Delphini. This strange inactivity shows that the systems 
it characterises are either of exceedingly small mass, or else 
inconceivably remote from the earth.^ 

We have positive knowledge of the masses only of such 
stellar couples as have had both their parallaxes (the equiva- 
lents of their distances from the earth) and their orbits 
determined. They are seven in number — namely, Sirius, 
a Centauri, 61 Cygni, 70 Ophiuchi, tj Cassiopei®, 85 Pegasi, 
and o* Eridani; and their total mass proves to be that of 
thirty suns.^ On an average, then, each of these systems con- 
tains nearly four and a half times as much attractive energy 
as the solar system, each individual star being more than equal 
in this respect to a pair of suns like ours. Were the extension 
of this mean conclusion legitimate, the distances of all stars 
revolving in ascertained orbits might be inferred from their 
assumed massiveness (since the relation between distance and 
mass is convertible), and upon this principle Madler derived 
what he called the 'hypothetical parallaxes* of binaries,.* 

* Madler, FixstemsysUme, p. 10. * See Appendix, Table V. 

' Dcr Fixstcrnhimmcli p. 82. 

o 1 


reckoning, however, the mass of each pair to be only that of a 
single sun. This estimate is now seen to be mnch too small, 
and the distances founded upon it to fall proportionately short 
of the truth.^ But, indeed, no general conclusions of the kind 
are fit for application to individual cases. The range of 
variety is so great that only simulated knowledge can be 
obtained in this way. Collective inferences, however, are not 
therefore worthless. Thus, from averaging the masses of 
only seven binaries, we have already gathered plausible 
grounds for believing our sun to occupy a low rank as a centre 
of attraction. It may be, nevertheless, that the swifter 
binaries, which can at present alone figure on such a list, 
give too high an average mass. 

Calculations based upon the orbital elements of revolving 
stars tell nothing of their relative masses. They apply only 
to the common stock of matter in each system, leaving its 
apportionment to be otherwise tested. This cannot be done 
except through the due apportionment of movement between 
the members of the system — an arduous task, hardly yet 
begun to be grappled with. 

There is no such thing in nature as a stationary body 
round which other bodies circulate. Answering motion there 
must always be, though on a scale reduced just in the same 
proportion that the mass is increased. Thus, the earth 
describes, under the influence of the moon, an ellipse exactly 
similar to that described by the moon under the influence of 
the earth, but eighty-one times smaller. And the sun corre- 
sponds in the same way to the revolutions of every one of the 
planets, notwithstanding that the centre of his movement as 
regards each of them, with the single exception of Jupiter, 
lies far beneath his own surface. Binary stars, however, are 
often probably almost equally massive, and therefore almost 
equally mobile bodies. The fixity of one member of each 
pair is purely conventional — an indispensable fiction without 

1 The masBes of revolving stars vary, cateris paribus, as the cubes of their 
distances from the earth. Of systems identical in period and apparent move- 
ment, one twice as remote as another will be eight times as massive, one three 
times as remote twenty-seven times as massive, and so on. 


which measurements would be impracticable. Those actually 
made give the sum of the movements of both stars, and an 
orbit computed from them represents the sum of their dis- 
tinct orbits. Identical in shape and position with the true 
ellipses, it differs from them only in size, its linear dimensions 
in any direction being equal to both theirs taken together. 

The genuine centre of movement of two mutually circling 
stars is their common centre of gravity, which Hes on a 
straight line drawn from one to the other, at a distance from 
each inversely proportional to its mass. The precisely similar 
orbits traversed by each are then spacious in the same inverted 
ratio. The larger star performs the smaller circuit, and vice 
versd. In the case of their equality, their orbits must inter- 
sect if elliptical, but coincide if circular, when the stars will 
pursue each other along the same track, while occupying in it, 


Fia. 30.— Orbits of the Components of y Virginis. 

at each successive moment, diametrically opposite positions, 
nor could either, during an eternity of undisturbed revolution, 
gain a hair's breadth upon the other. Circular movement 
ifl, however, rarely even approximated to in stellar systems, 
the members of which usually follow highly eccentric 

We may take 7 Virginis as an example of a pair moving 
in equal ellipses, thp relations of which are shown in fig. 80. 
They have, it will be seen, a common focus, the seat of the 
centre of gravity, from which the stars (being of equal mass) 
must always be equally distant. Neither can approach to or 
recede from this point of origin of the force acting upon them 
without the other simultaneously doing the same ; the two 
must be in periastron, and retire towards apastron together, 
losing, and subsequently regaining, velocity by the same grada- 
tions. The stars of a Centauri are also believed to travel in 


equal orbits, but in much less elongated ones than thoso of 
7 Virginia (see fig. 31). 

The movements of unequal stars are similarly conducted. 
That is to say, the proportion of their respective distances from 
the common focus is invariable. They are accordingly always 
found in corresponding parts of their orbits, and at opposite 
ends of a right line passing through the focus. The manner 
of their revolutions can be realised by a glance at fig. 82, re- 
presenting the orbits of Sirius and its companion, the small 
ellipse belonging of course to the bright star. 

Obviously, from what has been said, knowledge of the re- 
lative masses of binary stars would ensue upon knowledge of 
the relative dimensions of their separate orbits. But for its 
attainment, a prolonged scries of most delicate micrometrical 

Fig. 31.— Orbits of the Components Fio. 32.— Orbits of Sirius and 

of a Centauri. its companion. 

measurements between each of the revolving objects and some 
neighbouring star chosen as a reference-point, would in general 
be necessary.^ Such measurements could now be executed 
with comparative facility by photographic means, and their 
results can scarcely fail to afford valuable information regard- 
ing the physical condition of stars. The relations of mass 
to light-power, for instance, could thus be investigated with 
some definiteness. All that we know about them at present 
is that they vary largely, and often unaccountably. 

The proper motions of three binary systems have afforded 
some information as to the distribution of matter in them ; 
for the* track pursued by each component is necessarily a 
sinuous one, like that of the moon round the sun, while the 

* Pickering, Proc. Amer. Acad. vol. viii. p. 6 (1881). 


centre of gravity of the two advances uniformly in a straight 
line. Now this neutral point was found by Mr. Stone ^ to be 
situated about midway between the stars of a Centauri, show- 
ing them to be not far from equally massive ; it is just half 
as distant from Sirius as from its companion, and lies, accord- 
ing to M. Ludwig Struve,^ nearly five times closer to the large 
than to the small star of tj CassiopeisB. The latter object 
contains then f as much matter as its primary, yet it emits 
only -^^ of its light, so that it is notably deficient in bright- 
ness. The same peculiarity, in an enhanced degree, charac- 
terises, as we have seen, the Sirian satellite ; and its appear- 
ance in the secondary member of the great pair in the Centaur 
is the more remarkable that their equality in point of mass 
gives no good ground for the supposition of their having 
reached diJBferent stages of evolution. The key to the puzzle 
may be found in spectroscopic observations. 

From them, indeed, we may hope before long to get, not 
only interpretations of such results as we are now in search 
of, but the results themselves. For this purpose the prin- 
ciple of the change in the refrangibility of light through mo- 
tions of recession or approach can be made available. Altered 
refrangibility makes itself perceptible in displacements of 
known spectral lines, and these can now be measured on 
photographic plates ^ with such exactitude as to give perfectly 
trustworthy information respecting the radial velocities of 
the objects the displaced lines are derived from. If the ob- 
jects be the members of a binary system, the determination 
of their relative velocities in line of sight will sufiBce to fix 
their relative masses, since the proportion of their orbital 
movement, directed at any given instant towards or from 
the earth, must always be the same for each component. The 
inverted correspondence between mobility and mass thus holds 
good for radial measurements. 

These are facilitated by the circumstance that the simul- 
taneous ' end-on * motions of mutually revolving stars are, by 
the necessity of the case, oppositely directed. They reach a 

• Monthly Notices, vol. xxxvi. p. 258. 

' Copernicus, vol. i. p. 199. • See ante, p. 31. 


maximum when one is in the ascending, the other in the de- 
scending node of their orbits — that is, when one star is cross- 
ing the plane of projection on its hither course, the other on 
its further excursion ; they disappear on the stars reaching 
two intermediate positions, when the whole speed of circula- 
tion jflows transversely to the line of sight.* These alterna- 
ting changes make it possible, by repeated observations at 
sufficient intervals of time, to eliminate the constant effect 
due to the proper motion of the system. The variable residue 
will then be the element sought. 

Its determination promises to be fruitful in several ways. 
The total orbital velocity of each pair for which it was ascer- 
tained could easily be calculated, and the dimensions of the 
orbit thus implied would disclose the absolute mass and dis- 
tance from the earth of the combined stars.' Still a further 
piece of information would be conveyed. Simple visual ob- 
servations are helpless to discriminate between the nearer 
and the further sections of a stellar orbit. The effect to the 
eye is the same whether the moving star be approaching on 
the left-hand side and receding on the right, or approaching 
on the right and receding on the left of its apparent path. 
In other words, the ascending and descending nodes are in- 
distinguishable, and the computed ellipse may indifferently 
lie in the plane assigned to it, or in one equally tilted in the 
opposite dkection.^ This ambiguity might be removed by a 
single spectroscopic observation at either node, telling whether 
the star was advancing or retiring. 

A great number of binary couples will doubtless respond 
satisfactorily to this last simple test ; but only those most 
favourably situated can be expected to prove susceptible of the 
more refined measurements leading to detailed results. The 
spectroscopic method has naturally no bearing except on stars 
moving in known orbits ; nor is it likely to be successfully 
applied unless the planes of those orbits are more nearly level 
than perpendicular to the visual line, and the objects traversing 

» Niven, Monthly Notices, vol. xxxiv. p. 339 ; Palisa, Asir. Nach. No. 2941. 
2 Bambaut, Proc. R. Irish Acad. vol. iv. Ber. ii. p. GG3. 
» R. Falb, SiriitSy Bd. vi. p. 121. 


them are bright enough, and suflBciently far apart, to show 
distinct spectra. Their remoteness, provided these conditions 
be fulfilled, is of no consequence,^ spectroscopic motion-dis- 
placements being absolutely independent of distance ; and it 
may accordingly be possible to determine in this way stellar 
parallaxes so minute as to be otherwise insensible. 

The pair on the whole best suited for experiments of this 
nature is a Centauri. In April 1879 the radial part of the 
velocity relative one to the other of the components 
amounted (according to Mr. Rambaut's calculation)^ to eight 
miles a second — a quantity easily measurable with the Pots- 
dam photographic apparatus. The same element for Sirius 
will reach a value, in September 1890, of thirteen miles a 
second ; but here the spectrum of the secondary star is much 
too faint to be communicative as regards motion. Among 
couples promising to become so are 70 Ophiuchi, X Ophiuchi, 
and ^ UrssB Majoris ; their treatment, to be successful, must, 
however, be exceptionally skilful, patient, and delicate. 

A remarkable circumstance connected with the last-named 
system gives a special interest to the spectroscopic determina- 
tion of its translatory movement. The stars composing it 
appear to move in a progressively widening orbit — a visual 
effect, it is conjectured, of their continuous approach to the 
earth.* Only the approach should be improbably rapid to 
account for the observations, and it has not yet been shown 
to be present at all. Theoretically, every stellar pair must be 
in course of separation or closing-up according as it is ad- 
vancing or receding along the line of sight, and its parallax 
could even (in the abstract) be derived from the propor- 
tion borne by the change to the rate of motion causing it. 
But modifications produced in this way are not likely, for 
an indefinite time, to enter into the domain of practical 

Another indirect method, scarcely less remote from 
realisation, of establishing the distances from the earth of 

» Fox Talbot, Repwt Brit Assoc. 1871, p. 34. Pt. ii. 

* Proc, B. Irish Acad, vol. iv. p. 669. 

' Birkenmajer, Sitzungsberichte, Wien, Bd. xciii. il. p. 786. 


computed binaries, was proposed by Savary in 1827.* It aims 
at learning the real dimensions of stellar orbits from the 
time taken by light to cross them ; and this would become 
known through the delay in the visibility of the more remote 
component (simultaneous observations givingnon- simultaneous 
places),' if only acquaintance with stellar revolutions were 
indefinitely more accurate than it is. Thus the * equation of 
light ' for 61 Cygni, the circuit performed by which is the most 
spacious we are cognisant of, is only nine hours, an interval 
during which the orbital motion of the pair is utterly lost in 
the uncertainties of calculation. 

In an amendment to Savary's plan, Houzeau attempted 
in 1844 ^ to turn to account for the same purpose certain 
optical inequalities produced in the onward motions of con- 
nected stars by the progressive transmission of light. But 
their interest is almost purely speculative. Villarceau showed 
that the only possibility (and that a remote one) of their being 
rendered apparent depended upon the preliminary ascertain- 
ment of the relative masses of the stars concerned.* It is 
then safe to predict that this species of Might-aberration* 
will not prove instrumental for the advancement of definite 

Professor Pickering* and Mr. Monck® of Dublin have 
separately perceived the existence of a relation between 
the movements and magnitudes of binaries, rendering it 
possible to determine their comparative superficial luminous 
power quite independently of their distances. It is neces- 
sary, however, to assume either that the components of 
each pair are of similar quality in this respect, or else that 
one of them is of negligeable mass; nor does the formula 
distinguish between extent of surface and intensity of shining. 
The degree in which matter is concentrated in the bodies 

> Conn, dea Temps, 1830, p. 169 ; Arago, Annuaire, Paris, 1834, p. 286 ; 
Stnive, Mens. Microm. p. olxxii. 

» Birkenmajer, loc. cit. p. 717. • Aatr. Nach. No. 496. 

* Conn, des Temps, 1878, Additions, p. 3 ; Comptes Rendus, t. xxxiv. p. 353 ; 
Seeliger, VierUljahrsschrift Astr, Ges, Jahrg. xxi. p. 286. 

* Proc, Am, Acad, vol. viii. p. 14 (1881). 

■ Observatory, vol. x. p. 96 ; Knowledge, vol. xii. p. 141. 


considered is left untouched by it ; their mean density may 
have any assignable value. 

The result of these inquiries is to confirm the prevalence 
of astonishing varieties in the emissive powers of different 
stars. Thus, 7 Leonis is three hundred times more brilliant 
than the sun, if its density be the same ; or, if we suppose 
its brilliancy the same but its density inferior, then the 
substance of these globes must be seven times rarer than 
atmospheric air at standard pressure, in order to give the 
required bulk,* It is possible, however, that their move- 
ments, when better known, may disclose their possession of 
more gravitating power than Dr. Doberck's orbit implies. 
Between their discovery by Herschel in 1782, and 1878, they 
traversed an arc of only 85**, leaving obviously much room for 
uncertainty as to their future course. Their spectrum, which 
is of the solar type, in no way accounts for their abnormal 

In the same way, the brilliancy of S Cygni comes out (with 
Behrmann's elements) one hundred times that of the sun ; 
S Equulei and f Sagittarii emit fully thirty times, 7 Virginis 
and 25 Canum Yenaticorum more than sixteen times. Castor 
nearly fifty times the solar Ught relative to mean density. 
Indeed, our solitary star is so generally and so far surpassed 
in luminosity by the members of binary systems, that Pro- 
fessor Pickering inclined to think that injustice had been 
done to it in current estimations. Stars are nevertheless to 
be found standing vastly beneath it as light-givers. We have 
already met with one notable instance to this effect in the 
companion of Sirius, and there are several others. Thus 
about half the sun's brilliancy belongs to 61 Cygni, and no 
more than ^V P^* ^^ i* ^^ claimed by a pair of revolving 
ninth-magnitude stars forming part of the triple combination 
known as 40 (otherwise o^ Eridani. The singular contrast 
between the brightest and duskiest of known star-couples was 
pointed out by Mr. Monck. ' If 7 Leonis,' he remarks, * were 
replaced by a star of equal mass but similar in character to 
40 Eridani, its light would be reduced to ^^nr of its present 

* Proc, Amer, Aca4, vol. viii. p. 14, 


intensity, while if the fainter pair of 40 Eridani were re- 
placed by a star of equal mass whose brilliancy was eqnal to 
that of 7 Leonis, its light would be increased twenty thousand 
fold, and it would outshine every star in the sky except 

The most striking general peculiarity of stellar orbits is 
their high eccentricity. Nearly all of them are greatly more 
oval than the planetary paths round the sun, and a large 
proportion approach to cometary shape. This remarkable 
fact indicates, in Professor Kirkwood's opinion,* the division 
of the parent nebul» of double stars before the acceleration 
by contraction of their rotatory movement had made much 
progress. As regards the development of tangential velocity, 
these systems were thus in a sense abortive, and their mem- 
bers, beginning to revolve under the nearly unbalanced in- 
fluence of their mutual gravitation, necessarily pursued very 
much elongated tracks. There is no tendency to an agree- 
ment between them as to the direction in which these tracks 
are pursued. The revolutions of binary stars are indifferently 
retrograde or direct. Whether they to any extent affect a 
common plane is a question that has yet to be decided. If 
they do, their preference is not for the level of the Milky Way, 
but for that at right angles to it.' Nothing like proof, 
however, has as yet been given of the regulation, by any 
fixed principle, of the inclinations of stellar orbits. The 
existence of a fundamental plane of movement would be of 
high significance as regards the history and relations of the 
sidereal world; yet the whole drift of modern research 
suggests, rather than the close and rigid union between its 
parts which it would indicate, a loose connection destined to 
be extensively modified by time. 

* Journal Liv. Astr. Soc. vol. v. p. 174. 

* Atner, Jour, of Science^ vol. xxxvii. p. 233 (1864). 

* Gore, Eng. Mechanic, vol. xlvi. p. 296. 




The further resolvability of a great many double stars is 
perhaps the most curious result of modern improvements in 
the optical means of observing them. With every addition to 
the defining power of telescopes, the visible complexity of 
stellar systems has increased so rapidly as to inspire a sus- 
picion that simple binary combinations may be an exception 
rather than the rule. The frequency with which what 
appeared to be such have yielded to the disintegrating scrutiny 
of Mr. Biurnham and others, suggests at any rate the presence 
of an innate tendency, and seems to show that the duplicity 
of stars is no accident of nebular condensation, but belongs 
essentially to the primitive design of their organisation. 
Although we can never become fully acquainted with all the 
detailed arrangements of stellar systems^ we are then led to 
suppose thepi far more elaborate and varied than appears at 
first sight. Each, we cannot doubt, is adapted by exquisite 
contrivances to its special end, reflecting, in its untold har- 
monies of adjustment, the Supreme Wisdom from which they 

The continuance of the process of optical dissociation, 
begun by the splitting-up of an apparently simple star, some- 
times shows the primary, sometimes the satelUte, not unfre- 
quently both primary and satellite, to be very closely double. 
Ternary systems are accordingly of two kinds. In one, the 
smaller star consists of two in mutual circulation, and con- 
current revolution round a single governing body; in the 
other, an intimately conjoined pair guides the movements of 
an unattended attendant. The planetary type of construction 
is uncommon or unknown. No star has been ascertained to 


possess two or more companions circulating (so to speak) co- 
ordinately. Groups possibly indicating such a disposition of 
parts exist, but perspective may have a share in producing 
them. The variable S Monocerotis, with its two client-stars, 
is an example. 

Among the most interesting triple stars of the double- 
satellite description is the brilhantly coloured 7 Andromedae. 
The original components, of third and fifth magnitudes 10" 
apart, remain in statu qxio since they were seen by Father 
Mayer in 1777, but their secular journey together over an 
arc of &' establishes the genuineness of their relationship. 
In 1842 the sea-green companion was found . to be itself 
double. With the fifteen-inch Pulkowa refractor, Otto Struve 
caught sight of a thin black line (representing probably a gap 
of some thousands of millions of miles) dividing it into two 
stars, of respectively 6^ and 7 magnitudes, since perceived to 
be revolving at a rate which would carry them completely 
round their orbit in about 500 years. They have already 
progressed so far as to obliterate the * thin black line ' testify- 
ing to their separate existence. Even Mr. Burnham's keen 
eyesight, aided by the utmost powers of the Lick thirty-six 
inch, is no longer adequate to distinguish them. All that can 
be seen is that the blended object they constitute, instead of 
being perfectly round, like a truly single star, is slightly 
* oblong ' in one direction. When the pair was more open than 
it is now, a difference of tint between the components was 
noticed by several observers,* and it is not improbable that the 
emerald effect of their light results from the merging together 
of actually distinct yellowish-green and blue radiations. 

A pair in some respects similar, but much fainter, is 
attached at about the same apparent interval to the lustrous 
white star Eigel. An excessively diflScult object at the time 
of its detection, it has of late become impossible. Mr. 
Burnham has for some years failed to extract from it the 
slightest sign of duplicity.* Yet its elongation was suspected 
both by him and Mr. Herbert Sadler in 1871,^ and was 

» Gledhill, Observatory, vol. ii. p. 269 ; Bameby, 16. p. 229. 

2 Asir. Nach. No. 2929. ' ' A^r, Register, vol. xvui. p. 15. 


laboriously verified, first at Chicago in 1877, then from 
Mount Hamilton in 1879. Change, under these circumstances, 
seems much more probable than error, and great interest 
will attach to future observations of the tiny sapphire star 
joined in disparate union with the chief luminary of Orion.* 

A ternary group, corresponding to these two in plan, but 
greatly enlarged in optical scale, has already been cursorily 
noticed.* It consists of a 4*6 magnitude star, designated by 
Flamsteed 40, by Bayer o* Eridani, with a faint and far-away 
double satellite, all three discovered by Herschel in 1783. The 
physical association of the pair with the large star at an 
interval of 82" would be improbable, were it not certified by 
their possession in common of an exceptionally swift proper 
motion. An advance during the last century over a space 
nearly equal to a quarter of the moon's diameter has modified 
their relations only by a trifling approach to their primary of 
the dependent stars, due perhaps to slow circulation round it 
in an orbit presented edgeways to our sight. 

They have, in the mean time, almost finished a circuit of 
each other, and will have completely finished it, by Mr. Gore's 
calculation, within 139 years from the date of their detec- 
tion. And since their distance from the earth has been 
measured, the real size of their orbit and their joint mass 
are also known. We find then that the average interval 
between them is thirty-six times that separating the earth 
from the sun, so that (their path being only moderately eccen- 
tric) they never approach as near to each other as Neptune 
does to our central orb, which they together surpass two and 
a half times in gravitative power. But in their place, from 
which Ught reaches us in nineteen years, the sun would 
shine as a fifth-magnitude star, while they combine into one of 
only ninth magnitude. Their feeble luminosity thus once 
more forces itself upon our attention, and compels us to reflect 
upon the possibility of whole systems existing in unimpaired 
mechanical perfection, but wrapt in perennial darkness. For 
what purpose existing, who can tell? The flight of our 

* Bigel, although oalled/3 Orionis, decidedly sarpasses a Orionis in brightnesB. 
« Bee ante, p. 203. 


thoughts is short, and the ultimate aims of the Maker are re- 
mote. To attempt to compass them is to invite palpable failure. 

Among double primaries Castor holds the first place, 
through the lustre of its components and the vastness of the 
scheme completed by the captive star borne along in their 
train. Another specimen is s Equulei, one of Herschel's 
pairs, the larger member of which was again divided by Struve 
in 1835. The feat had become possible through the progress 
of orbital motion, the continuance of which has since rendered 
it easy. Signs of circulation in the 7 '6 magnitude star at 
11" perhaps exist,* but inconspicuously. It is, however, an 
undoubted satellite of the close couple. 

Physical, too, almost certainly is the group constituting 12 
Lyncis. Here a white star of the sixth magnitude has an 
immediate neighbour nearly as bright, but reddish, the two 
describing in 486 years orbits sUghtly less eccentric than that 
of the minor planet Pallas.' Unless their joint mass be less 
than the solar, they must be so far off that their light takes 
at least 120 years to reach our eyes. Their brightness relative 
to mass is twice that of the sun. Considering the length of 
their period, the fixity of a bluish attendant at nearly five 
times the mean interval between them is not surprising. 

The movements of the third star (of 7*2 magnitude) in 
the ternary combination f Scorpii are of a somewhat per- 
plexing nature. They suggest interruptions, due possibly to 
disturbances by an unseen agency. Madler conjectured for 
them a period of 1469 years;' a couple of centuries may, 
however, elapse before they develop suflficiently to be computed. 
They seem, too, to progress in an opposite direction from those 
of the close double star which controls them at an apparent 
distance of 7'' ; but we can learn only from the spectroscope 
whether this is really the case. 

The primary in this system consists of a fourth and a fifth 
magnitude star at l''-8, just separable, accordingly, with a 
good four-inch telescope. The orbit assigned to them by 

* Flammarion, Catalogue^ p. 139. • Gore, Asir, Nach, No. 2808. 

■ Tarrant, Journal Liverpool Astr. Society ^ vol. ▼. p. 205 ; Crossley, Hand- 
book, 320 ; Schorr, Observatory, August, 18U0, p. 281. 


Dr. Doberck approximates, in an unusual degree, to a circle, 
and is traversed in 96 years. The * mass-brightness ' of these 
objects is twelve times that of the sun, and they, no less than 
their attendant, are believed to be slightly variable. Their 
spectrum is of the Sirian type. 

The relations of the three stars of jS Scorpii are diflFerent, 
but equally genuine. Two, of respectively third and fifth 
magnitudes, but almost certainly variable, were first observed 
by Herschel in 1779; the third, discovered by Burnham a 
century later, makes with the primary an exceedingly difl&cult, 
unequal pair at 0"*96.* No relative movements have yet 
been perceived in this system, which nevertheless asserts its 
organic unity by the harmony of its advance through space. 

One of the most curiously interesting of all the stellar 
systems known to us is ternary from a visual, quaternary 
from a physical point of view. It is composed of three bright 
members and an obscure one, all in comparatively rapid mutual 
circulation. The division of f Cancri, by Tobias Mayer in 
1756, into a fifth and a sixth magnitude star about 5i'' asunder 
was the preliminary to Herschel's further analysis. *If I 
do not see extremely ill this morning,' he wrote on Novem- 
ber 21, 1781, 'the large star consists of two.'^ This was the 
earUest example of the decomposition of a double into a triple 

The next distinct view of these close objects (called for 
convenience A and B, the remoter star C) was obtained by 
Sir James South at Passy in 1825, but Struve's nine-inch 
Praunhofer showed them easily, and they have never since 
been lost sight of. Re-observation at once rendered patent 
their swift movement of revolution. Before the close of 1840 
they had, by resuming the positions in which they were 
originally observed, authoritatively declared their period to be 
not far from sixty years. And their orbit lies in a plane so 
nearly square to the line of sight, that foreshortening takes 
little eflFect upon it, and occultations are hence impossible. 
Although the maximum interval between the stars scarcely 
exceeds one second, and the minimum interval is no more 

• Monthhj Notices, vol. xl. p. 100. * Crossley, Handbook, p. 247. 



than 0"-2 ; they never closfe up beyond the dividing powers of 
first-class instruments. 

But the orbital movements of the couple A B make only 
part of a complex scheme of displacements. * This star/ Sir 
John Herschel remarked in 1826, 'presents the hitherto 
unique combination of three individuals, forming, if not a 
system connected by the agency of attractive forces, at least 
one in which all the parts are in a state of relative motion.'* 
And he added that, if really ternary, its perturbations must 
present 'one of the most intricate problems in physical 
astronomy ' — a forecast which bids fair to be fully verified. 

The star C apparently retrogrades round A B at an average 
rate of half a degree a year, indicating (if maintained with 
approximate uniformity) revolution in a period of 600 or 700 
years. But this average rate is subject to very remarkable 
irregularities. The path traced out in the sky, far from being 
a smooth curve, is looped into a series of epicycles, in travers- 
ing which the star alternately quickens and slackens, or even 
altogether desists from its advance, while increasing or dimi- 
nishing, by proportionate amounts, its distance from the centre 
of motion. This anomalous behaviour, detected by M. Flamma- 
rion in 1873,^ was both detected and interpreted by M. Otto 
Struve in 1874.^ The vagaries of the third component of 
f Cancri proved, from his investigation, to be very far from 
immethodical. The accelerations which they included were 
shown to be perfectly compensated by retardations, and to be 
accompanied unfailingly by expansions outward of the parts 
of the track where they occurred, while contractions inward 
attended slackened movements. An explanation too was 
hazarded, the substantial truth of which was amply attested 
by M. Seeliger's subsequent researches.* 

It seems then that the star C is merely a satellite to a 
dark body round which it describes, in 17^ years, a little 

> Phil. Trans, vol. cxvi. p. 326. 

' Catalogue^ p. 49. ■ Comptes Rendus, t. Ixxix. p. 1463. 

* Sitzungsherichtc, Wien, Bd. Ixxxiii. Abth. 2, p. 1018 ; Denkachriften^ 
Munich, Bd. xvii. Abth. i. 1889 ; Harzer, Astr. Nach. No. 2764 ; Observatory, 
vol. xii. p. 116. 


ellipse with a mean radius of one-fifth of a second. Together 
this singular pair circuits, or, more probably, is circuited by 
A B, the invisible disturbing body being, quite possibly, the 
most massive of the system. If this be the case, it is also, of 
course, the most nearly stationary, and should be regarded as 
the centre round which the lucent trio revolve — an arrange- 
ment hinting to us that the collocation in the same orb, 
familiar to us in the solar system, of the functions of rule and 
Ught-giving may, on occasions, be dispensed with. An anti- 
Copemican system, at any rate, appears to be to some extent 
exemplified by ? Cancri. Here a cool, dark globe, clothed pos- 
sibly with the vegetation appropriate to those strange climes, 
and plentifully stocked, it maybe, with living things, is waited 
on, for the supply of their needs, by three vagrant suns, the 
motions of which it controls, while maintaining the dignity 
of its own comparative rest, or rather of its lesser degree of 
movement. For the preponderance of this unseen body cannot 
approach that of a sun over its planets; hence its central 
position is by no means undisturbed. 

Another lesson already learnt from the stars of a Centauri ^ 
is emphasised by the relations as to mass and luminosity of 
the components of ? Cancri, It is, that such relations are not 
merely prescribed by the inevitable progress of cooling. For, 
if they were, the most ponderous should, among bodies of 
contemporaneous origin, be invariably the least advanced; 
it should still be a distributor of light and heat after all its 
companions had sunk to extinction. But in f Cancri the 
largest reservoir has been the soonest exhausted ; the globe 
containing most matter has far outstript its associates, and 
reached the planetary stage while they are still in their meridian 
glory as Sirian suns. The contrast is heightened in the close 
pair, by their possession (according to Professor Pickering's 
calculation) of nine times the solar emissive power relatively 
to mass; and all the three visible components are in this 
respect most Ukely of all but identical quality. Their real 
differences of magnitude, too, seem to be slight, although at 
times exaggerated by relative variability. The entire group 

' Se3 ante, p. 199. 

V 2 


is transported across the sphere at the rate of 15" a century, 
but its distance from the earth is unknown. 

A quadruple system of an unique kind is perhaps formed by 
f Urs8B Majoris with three variously related attendants. Already 
mentioned as a slowly revolving double star,* it is besides 
escorted on its indefinite journey onward by the fifth-magni- 
tude star Alcor, the two making the combination popularly 
designated the * Horse and Eider.' Since the interval between 
them is of 11 '30", they can easily be distinguished with the 
naked eye ; nevertheless Alcor, totally overlooked by the 
Greeks, was regarded as a test-object for keen eyesight by 
the Arabs. Its gradual brightening is thus strongly sug- 
gested.' The probability that Mizar and Alcor mutually 
revolve is strong, but not overwhelming; their connection 
might be otherwise explained.* If they do, their anntcs inagjius 
must be of enormous, to our ideas of interminable length. An 
estimate for it of 190,000 years is, however, purely conjectural. 

The fourth member of the group has betrayed its existence 
after an absolutely unprecedented fashion. On the Harvard 
photographs of the spectrum of Mizar, the lines contained in it 
come out periodically doubled^ This can only mean that two 
approximately equal stars are united into a single telescopic ob- 
ject so closely as to be distinguishable only through the line-dis- 
placements due to their orbital revolution. Since one compo- 
nent must always retire from, as the other approaches the earth, 
the lines in their respective spectra are at such times pushed 
in opposite directions, and imprint themselves consequently in 
duphcate upon the sensitive plate ; when, on the other hand, 
the stars are moving across the visual ray, the lines cease to 
be shifted and appear single. This naturally occurs twice in 
each revolution, and those of f Ursse are in this way found 
to be accomplished in 104 days, with a velocity, oppositely 
directed for each star, of about fifty miles a second. They 
are hence 148 millions of miles apart, or about as far as Mars 
is from the sun, and their total mass equals that ot forty suns. 

* See antey pp. 168, 195. ' Flammarion, Catalogue^ p. 75. 
« See infra, p. 347. 

* Fourth Annual Report, p. 6 ; Pickering, Atnerican Jour, of Science, vol. 
xxxix. p. 46. See Monck's anticipation of this result in Jour. Liv. Astr. Soc. 
vol. vi. p. 115. 


Unless, indeed, their orbit be inclined to the visual ray from 
the earth ; in which case, both its size and their mass should 
be increased proportionately to the amount of the inclination. 

This unexpected discovery has been followed up by the an- 
nouncement that the spectral lines of y9 Aurigae are doubled 
every alternate night, implying orbital revolution in a period 
of four days. We have then plainly arrived at the threshold of 
a new era as regards the investigation of close stellar systems, 
and can only await in silent wonder the development of their 
strange peculiarities. The varied possibilities meanwhile 
of stellar companionship are exemplified by VogePs spectro- 
graphic discovery that Spica (a Virginis) revolves at a minimum 
rate of fifty-six miles a second, in an orbit with a radius of 
• three million miles, round the common centre of gravity of 
itself and an obscure, or partially obscure, companion.* His 
suspicion that Rigel is similarly coupled may have been con- 
firmed by the time these lines meet the eye of the reader. 

Eeal quatefnary stars are often self- discriminating ; their 
arrangement into two adjacent couples is more character- 
istic of physical connection than any possible distribution 
of thfee stars can be. And in effect, several perspective 
groups of a single star with a genuine pair, such as SEquulei, 
85 Pegasi, and j3 Delphini, are visibly in course of being dis- 
solved by proper motion, while no ' double-double ' combina- 
tion has yet given signs of breaking up. 

A representative specimen of the latter class offers itself 
in 6 LyrsB, a star of the fourth magnitude, a little to the 
north-east of Vega. Exceptionally keen eyes show it as 
double, and one of the brilliant surprises provided by the 
heavens for Sir William Herschel was that of finding each 
component further divisible. The discovery, though beautiful 
and interesting, was easy ; all the four stars can be seen with 
a good three-inch telescope. The * preceding ' pair, or that 
which crosses the meridian first, is distinguished as s^ the 
' following ' pair as e^ Lyrae ; and Flamsteed attached the 
numbers 4 and 6 to them respectively. The former consists of 
a fifth and a sixth magnitude star 3'' asunder ; the constituent 

* SiteungsberichtCf Berlin, April 24, 1890. Cf. Maunder, Observatory, July 
1890, p. 238. 


stars of s^ are nearly equal (5*8 and 5*5 magnitudes), and are 
set a little closer together (at 2''-46). Their revolutions, too, 
appear to be performed about twice as quickly as those of the 
neighbouring couple. From the shifting of their relative 
situations since 1779 to the extent of nearly half a right angle, 
their period may be estimated at about one thousand years ; 
while that of e* Lyree can hardly fall short of two thousand. 
The practicability of computing either orbit is still remote. 

The small common proper motion (9'' a century) of these 
bright couples affords positive evidence of their union into 
one vast system. At their unmeasured, perhaps iromeasur- 
able distance, the gap between them of 8^'' may well stand 
for a chasm costing light itself some months to bridge ; yet 
the stress of their mutual gravity reaches across it, compelling 
their circulation in orbits so spacious that a single round of 
them must occupy an era of no insignificant duration, even 
in the life of a star. The four stars of s LyrsB give a spectrum 
of the first type, combined, in the first two, with a decidedly 
yellowish colour. But this is often the case with double stars. 
* A miniature of s LyrsB ' ^ is offered to our regard in 
V Scorpii. This is perhaps the most beautiful quadruple gi'oup 
in the heavens, from the narrow limits within which the bril- 
liant objects composing it are crowded. As a wide double it 
was noticed by C. Mayer in 1776 ; after seventy years the 
smaller star was divided by Mitchel at Cincinnati ; and the 
larger one of fourth magnitude yielded similarly, in 1874, to 
the insistence of Burnham. These last very close stars are 
certainly revolving,^ and both pairs, at an interval of 48'', 
share a slight drift through space. 

The sixth-magnitude star 86 Virginis may be said to consist 
of a double primary with a double satellite at 27''. Acquaint- 
ance with the group in its true aspect was made through Bum- 
ham's analysis of one of Struve's wider pairs. The movements 
doubtless progressing in it have not yet become perceptible. 

Eighteen * double-double ' combinations (one (2 2486) with 
a total extent of no more than 15") were enumerated by 

* Flammarion, Cataloguet p. 96. 

* Burnham, Memoirs B, Astr. Sockiy, vol. advii. p. 288. 


Barnham in 1882,^ and two were subsequently discovered by 
Professor G. W. Hough." They perhaps exist more nume- 
rously than we have as yet any idea of. 

A ' double-treble ' star, so called by Herschel, has been the 
subject of numerous successive discoveries. With the slight- 
est optical assistance a Orionis, a star of 3*7 magnitude, just 
beneath the middle star in the belt of Orion, separates into 
two wide and unequal components, each of which was, Octo- 
ber 7, 1779, perceived by Herschel to be triple.* As usual in 
such cases the process of resolution was continued, and the 
assemblage was described by Barlow as ' double- quadruple, 
with two very fine stars between the sets.* * These last, how- 
ever, are not unlikely to be mere optical associates. To this 
intricate group Burnham has added a further element of 
complexity by detecting, in the autumn of 1888, its leading 
member* to be formed of a fourth and a sixth magnitude star, 
a quarter of a second apart.' The discovery, like some others, 
raises a question as to the point where stellar subdivision can 
really be said to cease. Certainly not where visual limitations 
interfere with our recognition of it. 

The essential character of <t Orionis is that of being made 
up of two distinct, yet evidently connected hwU of stars, and 
the same knot (2 672) ® contains all the four brightest com- 
ponents. These differ, and perhaps vary in colour, and their 
influence may be assumed to predominate in a system, which, 
however, gives no sensible evidence of movement, whether 
of the circulatory or the advancing kind. 

The multiple star 45 Leporis is organised on a plan less 
markedly definite than that governing the structure of 
<7 Orionis. It consists of four principal and five subordinate 
members, the last successively discovered by Sir John Herschel, 
Burnham, and Hall.^ The star B, of eighth magnitude, stands 
out through its ruddy colour from its white companions.® 

' Observatory, vol. iv. p. 176. * Astr. Nach. No. 2778. 

» Phil Trans, vol. Ixxii. p. 124. 

« Smyth, Cycle of Cel Objects, ed. 1881, p. 156. * Astr. Nach. No. 2875. 
• Struve, McTis. Microm. pp. 149, 245. 

' Burnham, Memoirs It. Astr. Soc. vol. xliv. p. 238 ; Astr. Nach. No. 2062. 
Observatory, vol. iv. p. 177. » G. Knott, ibid. pp. 184, 212. 


Measurements of the groups are still so recent that their 
repetition can hardly be expected for some time to come to 
give evidence of motion. Blended into a single object, the 
nine stars, covering an extent of 125", are just visible to the 
naked eye. 

The nebular relations of double and multiple stars were 
noticed with surprise by Sir John Herschel at the outset of his 
career.* Although admitting without hesitation their physical 
character, he was without the means of establishing it since 
made available, and could support his conWction only by the 
utter improbability of such collocations as he pointed out being 
fortuitous. Thus, a close, mijjute stellar couple is planted at the 
exact centre of a faint round nebula in Leo (New Gen. Cat., No. 
8230) ; and the same kind of coincidence recurs twice in the 
southern constellation Dorado (N. G. C, Nos. 1732, 1951). Two 
pairs in Sagittarius, each set in the midst of a nebula (N. G. G., 
Nos. 6589, 6590), may from their contiguity be suspected to 
constitute one system ; and two ninth-magnitude stars at 15'', 
marking very nearly the foci of an elliptical nebula in the 
same region (N. G. C.,No. 6595), are certainly not accidentally 
projected upon it. * One of the most curious objects in the 
heavens' (according to Sir John Herschel),* is a trio of stars 
arranged in an equilateral triangle, the sides of which measure 
4", and relieved upon a shield of milky light Qi. G. C., No. 
1931) ; an analogue, possibly, of two adjacent stellar trios 
observed at Harvard College in 1860, in the places of two 
small but bright nebulae recorded by Bond and Auwers in 
1853 (N. G. C, Nos. 2399, 2400). Mere dififerences of seeing 
might, it is true, account for these apparent substitutions ; 
but the alternative supposition that the stars came more 
plainly into view through a loss of light in their nebulous 
surroundings deserves at least to be tested by renewed ex- 
periments, visual or (better still) photographic. But there 
are other still more noteworthy instances of the association 
of composite stars with nebulae. The whole framework of the 
great nebulous structure in the sword of Orion seems to rest 
upon the stellar group designated 0, or rather 5' Orionis ; for 

' Memoirs i?. Astr. Soc. vol. vi. p. 78. '• Ibid. vol. iii. p.54. 


there is a second 6 not far oflf, itself a wide double star, and 
the two together form, to the eye, one diffuse object, singly 
catalogued by Ptolemy, Tycho Brahe, and Hevelius. But it is 
with 6* exclusively that we are at present concerned. 

On the very slightest telescopic persuasion, it allows itself 
to be seen as quadruple. The four stars into which it divides 
are severally of fifth, sixth, seventh, and eighth magnitudes, 
the greatest interval between any two of them not exceeding 
21''. None of them is in visible subordination to any other ; 
they stand, it might be said, on an equal footing, at the four 
corners of a rudely quadrilateral figure, or * trapezium.' They 
maintain their places, too, both absolute and relative, with 
singular rigidity. After two and a half centuries of observa- 
tion, no shifting of them can be detected. They are hence 
likely to be at a prodigious distance from the earth. 

The rule that such groups seem more crowded as they are 
better seen, has not been infringed here. A fifth star of the 
eleventh magnitude was added to the company by Struve, 
November 11, 1826, and a sixth, then still fainter, but which 
has since probably gained somewhat in light, by Sir John 
Herschel, February 13, 1880. Both of these, though closely 
associated, each with one of the larger stars, share their 
apparent immobility.* Variability in light has often been 
ascribed, and as often denied to them. Burnham's experience 
is against it ; yet the curious fact that Robert Hooke saw the 
fifth star in 1664 with a non-achromatic three-and-a-half-inch 
telescope,* is strongly indicative of temporary brightening. 
Professor Holden at least found it undiscernible at Washington 
with so much of the great refractor left uncovered ; ^ neverthe- 
less, all six of the trapezium-stars have, at favourable moments, 
been made out with both achromatics and reflectors of three 
to four inches aperture.* It can scarcely then be claimed for 
Hooke's observation that it demonstrates change. 

* Bomham'B measures seem decisive on this point. See Memoirs R, Astr. 
Soc, vols. xliv. pp. 203, 237, xlvii. p. 244 ; Monthly Notices, vol. xlix. p. 367. 

* Micrographiay p. 242. 

* Washington Observations for 1877, App. ii. p. 7. 

* Webb, Cel Objects, p. 367. 


Further members of this group have, at various times, 
been half seen, half surmised ; but their existence, always 
problematical, has been disproved through the application of 
the Lick thirty-six inch ; for the three new stars perceived 
from Mount Hamilton by Alvan G. Clark and Barnard could 
certainly have been detected with no less powerful instrument. 
Two of them lie within the trapezium ; the third, a double 
star of extraordinary minuteness and difficulty, just outside it.* 
Their positions are shown on the accompanying diagram 
(fig. 88), where the names of the observers are attached to the 
recently discovered stars. 

About half a degree to the north and south respectively of 
6 Orionis are situated the double star C, and the triple i 

Barnard '^ 



Alvan Clark 
A • 

• E 

• B 
Fzo. 88.— stars of the Trapezium. 

Orionis, each with a considerable encompassing nebula, in 
manifest dependence upon the great intermediate mass 
gathered about the trapezium. But if all three (as appears 
certain) belong to the same formation, the stellar gi'oups con- 
stituting their nuclei must be connected, however remotely, 
into a single system. We cannot indeed regard them as per- 
forming mutual revolutions, which should be quickly arrested 
by the resistance of the gaseous stuff with which that whole 
region is filled ; and we are equally unprepared to admit their 
abandonment to the influence of gravity issuing in cata- 
strophic collisions. But we know of no third alternative. 
What then are we to conclude ? Only, it seems to us, that it 

> Bumham, Monthly Notices, vol. xlix. p. 352 ; Asir, Nach. No. 2930. 


is premature to conclude anything. The hint, however, as to 
the activity in the sidereal system of forces totally alien to our 
experience should be carefully borne in mind. 

In one other great nebula a multiple star is apparently 
dominant. The nuclear group in the trifid nebula (N. G. C, 
6514) consists of a close quartette covering an angular extent 
of only 19'', with two extremely faint additional stars dis- 
covered by Professors Langley and Holden. Complete apparent 
fixity characterises the arrangement. 

The frequent association of compound stars with nebulsB 
is no mere isolated fact. For they pass by insensible degrees 
into star-clusters, the fundamentally nebulous nature of which 
is rapidly, with the aid of photography, becoming established 
as an indisputable truth. The conjecture is even plausible 
that the formation of a multiple star in a great nebula repre- 
sents the initial stage of the development from it of a crowded 
cluster, minor nebulae giving rise to lesser groups ; and if 
objects of the kind have not yet, so to speak, been turned out 
of the workshop, it is no wonder that fragments of their raw 
material still cling round them. Compositeness of structure 
may thus measure primitiveness of condition, illustrating, 
though to us dimly, the sequence of Divinely decreed changes 
by which cosmical order is gradually more and more fully dis- 
engaging itself from the ' loud misrule ' of chaos. 




From multiple stars the transition is easy to star-clusters. 
These seem to embody completely the idea contained in germ 
in the former class of objects. They are collections, often on 
the grandest scale, of sunlike bodies small and large, united 
in origin and history, acted upon by identical forces, tending 
towards closely related ends. The manner and measure of 
their aggregation, however, vary widely, and with them the 
cogency of the evidence as to their organic oneness. There 
are innumerable cases in which it absolutely excludes doubt ; 
there are some in which it is rather persuasive than con- 
vincing. It is not then always easy to distinguish between 
a casual ' sprinkle ' of f^tars and a genuine cluster. Nor can 
the movement-test, by which so many physical have been 
discriminated from optical double-stars, be here applied. 
Internal displacements of a circulatory character have not 
yet become apparent in any cluster, and there is only one 
with an ascertained common proper motion. 

This is the immemorial group of the Pleiades, famous in 
legend, and instructive, above all others, to exact inquirers — 
the meeting-place in the skies of mythology and science. 
The vivid and picturesque aspect of these stars riveted, from 
the earliest ages, the attention of mankind; a peculiar 
sacredness attached to them, and their concern with human 
destinies was believed to be especially close and direct. Out 
of the dim reveries about them of untutored races, issued their 
association with the seven beneficent sky-spirits of the Vedas 
and the Zendavesta,* and the location among them of the 
centre of the universe and the abode of the Deity, of which 

• Bunsen, Die Pleiadm und der Thicrkreis, p. 434. 


the tradition is still preserved by the Berbers and Dyaks.* 
With November, the * Pleiad-month,' many primitive people 
began their year ; * and on the day of the midnight-culmina- 
tion of the Pleiades, November 17, no petition was presented 
in vain to the ancient kings of Persia ; * the same event gave 
the signal at Busiris. for the commencement of the feast of 
Isis, and regulated less immediately the celebration connected 
with the fifty-two-year cycle of the Mexicans. Savage Aus- 
tralian tribes to this day dance in honour of the * Seven Stars/ 
because 'they are very good to the black fellows.' The 
Abipones of Brazil regard them with pride as their ancestors.'' 
Elsewhere, the origin of fire and the knowledge of rice- 
culture are traced to them. They are the-' hoeing-stars ' of 
South Africa,* take the place of a farming-calendar to the 
Solomon Islanders, and their last visible rising after sunset 
is, or has been, celebrated with rejoicings all over the southern 
hemisphere, as betokening the 'waking-up time' to agri- 
cultural activity. 

To the Greeks of Hesiod's age their ' heliacal rising ' (the 
first visible before sunrise) announced, each May, the opening 
of the season for navigation ; and their name thus came to 
be interpreted (from plein, to sail) the ' sailing-stars.' But 
this etymology was doubtless— like the derivation of 'elf 
and * goblin ' from Guelf and Ghibelline — an afterthought ; 
and it may be confidently maintained that the word ' Pleiades,' 
bearing, like its Arabic and Hebrew equivalents, the essential 
signification of a 'cluster,' came from the Greek pleiones, 
many, or pleios, fuU.^ 

The similarity of the traditions respecting the swarm of 
celestial ' fireflies,' 

Quse septem dici, sex tamen esse solent, 
is as surprising as their universality. That they ' were seven 

* Haliburton, Nature^ vol. xxv. pp. 100, 317 ; Van Sandiak, L^ Astronomic,, 
t. iv. p. 367. 

* Haliburton, Festival of the Dead, p. 46. • Ibid. p. 13. 

* Lubbock, Origin of Civilisation, p. 316, 4th ed. 

* J. Hammond Tooke, in an interesting paper read in January 1^89 before 
the S. African Philosophical Society. 

* Nature, vol. xxxv. p. 608. 


who now are six/ is asserted by almost all the nations of the 
earth, from Japan toNigritia, and variants of the classical story 
of the* lost Pleiad* are still repeated by sable legend-mongers in 
Victoria, by * head-hunters ' in Borneo, by fetish-worshippers 
amid the mangrove swamps of the Gold Coast. An im- 
pression thus widely diflFused must either have spread from a 
common source, or originated in an obvious fact ; and it is 
at least possible that the veiled face of the seventh Atlantid 
may typify a real loss of light in a prehistorically conspicuous 
star. Some members of the collection are at present, there 
is little doubt, sHghtly or slowly variable,* and progressive 
tendencies of the kind are in more than one case suggested to 
be present. Thus Alcyone, the chief of the collection, now 
of the third magnitude, and just twice as bright as the 
brightest of its companions, was either not one of the four 
Pleiades observed by Ptolemy, or was then much fainter than 
it has been from Tycho Brahe's time to our own. So at least 
Francis Baily concluded from a careful examination of the 
records," and he knew better than most men how large an 
allowance has to be made for ancient inaccuracy. Abdurrah- 
man Sufi, too, the competent reviser of Ptolemy's observations, 
expressly states that the Alexandrian quartette appeared to 
him, in the tenth century, the most lustrous among the 
Pleiades.' Yet none of them can be identified with the 
actual lucida. A literal explanation of the old legend appears 
then feasible, and Professor Pickering's suggested identiJ&cation 
of Pleione with the missing Atlantid has much to recommend 
it.* The display by this star of a gaseous spectrum resem- 
bling that of P Cygni makes it at least fully credible that, 
like P Cygni, it should have been noted for temporary brilliancy. 
It is now of 6-2 magnitude — that is, just beyond the range 
of ordinary eyesight. 

The five stars ordinarily visible besides Alcyone (see 

* C. Wolf, Annales de V Observatoire de Paris^ t. xiv. ii. p. 26 ; Lindemann, 
Mimoires de VAcad,, St. P^tersbourg, t. xxxii. S^r. vii. No. 6. p. 29. 

* Memoirs R. Astr, Soc. vol. xiii. p. 9. 

' Schjellernp, Description des Etoiles, p. 132 ; Flammarion, Les ^toilest 
p. 294. 

* Astr. NacK No. 2934. 


Plate I. Frontispiece) are Electra and Atlas, each fluctuating 
slightly above and below 8*8 magnitude; Maia, now of the 
fourth, but on the rise ; Merope and Taygeta, the inferiors of 
Maia by respectively a quarter and a half magnitude. Celaeno, 
the seventh or concealed star, gives only about half as much 
light as Taygeta. 

It can be seen, however, with many others, under favour- 
able circumstances. Maestlin, the tutor of Kepler, perceived 
fourteen, and mapped eleven Pleiades previously to the inven- 
tion of the telescope ; Carrington and Denning counted 
fourteen,^ Miss Airy marked the places of twelve with the 
naked eye.^ The faintest of these fell but little short of the 
sixth, and there are twenty-five Pleiades down to the seventh 
magnitude,* each of which (with perhaps one or two excep- 
tions) might be separately visible in a transparent sky or 
from an elevated station. But their, crowded condition makes 
this impossible, and gives rise rather to the effect as described 
by Kazwini in the thirteenth century, of 'six bright stars 
with a number of dusky on«s between.'* 

With the use and increase of telescopic powers, the popu- 
lousness of the cluster has been amazingly increased. Aa 
object-glass scarcely exceeding two inches diameter showed 
Eobert Hooke in 1664 seventy-eight Pleiades,* and Michell's 
conjecture, in 1767, that there might be more than a thou- 
sand of them,^ has been superabundantly verified by the 
results of modern labours. Over an area about Alcyone mea- 
suring 1S5' X 90', M. Wolf catalogued, at the Paris Observa- 
tory in 1876, 625 stars to the fourteenth magnitude ; on the 
MM. Henry's sensitive plates, in 1885, 1,421 made their appear- 
ance in a smaller space, and the number was brought up to 2,326 
by exposures of four hours in November and December 1887. 
The faintest objects thus registered, although nominally of 
the eighteenth, were probably, by strict photometric reckoning, 
of about the sixteenth magnitude. 
- How many of them really belong to the group, and how 

» Webb, Cel. Objects, p. 393. * Monthly Notices, vol. xxiiLp. 175. 

* Harvard Annals, vol. xiv. pt. ii. p. 398. * Ideler, Stemnamen, p. 147. 

* Micrographia, p. 241. • Phil. Trans, vol. Ivii. p. 259. 


many are referred to it by perspective, can be determined 
with the help of time and patience. As regards some of 
the better-known stars, the process of discrimination has 
akeady begun. 

BesseFs measurements of the places, relative to Alcyone, 
of 52 Pleiades,' executed with the Konigsberg heliometer during 
the twelve years from 1829 to 1841, furnished a starting-point 
for investigations of their internal movements. The upshot 
of the first effective comparisons was to exhibit these as null. 
From a collodion-print of the cluster taken by Rutherfurd of 
New York in 1865, Dr. Gould redetermined nearly all Bessel's 
stars with such accuracy as to make it certain that no 
appreciable interstitial shiftings had occurred in the course of 
a quarter of a century.^ 

Now this seeming rigidity in effect implied a great deal. 
For the point of origin of the measures in question is not 
immovably fixed in the sky. The chief Atlantid has a secular 
proper motion (according to Newcomb) of 5"-8, the possession 
of which in common by the whole stellar band virtually de- 
monstrated their effectual union. Where one among many 
objects is ascertained to be moving, relative fixity can only 
mean that all drift together ; and so the unique phenomenon 
was brought to light of the transport in block across the 
sphere of a couple of thousand congregated suns. Even if 
the whole of this apparent displacement should prove to be, 
as it were, reflected from the solar advance, its significance 
of physical kinship among the objects affected by it would be 
nowise impaired. For the circumstance of their being simi- 
larly affected by it would suffice to locate them in the same 
region of space under the immediate influence of their con- 
straining mutual gravity. 

The establishment of a general unanimity of movement 
among the Pleiades was the first step towards investigating 
their relations ; the next was to seek evidence of systematic 
change. This perhaps remains to be found ; its highly recon- 
dite nature has at least been rendered unmistakable by the 

' Astr, Nach, No. 430. - Observatory, vol. ii. p. 10. 


labours of Wolf/ Pritchard,* and Elkin.* Displacements within 
the cluster, if they have been genuinely measured at all, are 
barely nascent. But with its analysis some progress has been 
made, as the result of Dr. Elkin's work at Yale College in 

Leaving nothing to be desired in the way of skill and care, 
it was the more strictly comparable with Bessel's from having 
been executed, like his, with a heliometer, one of exquisite 
workmanship, completed in 1882 by the Messrs. Eepsold, of 
Hamburg. Sixty^^nine stars, down to 9*2 magnitude, were 
included in the survey, only one of Bessel's being omitted, 
while seventeen were added from the Bonn Durchmusterung. 
The close agreement, on the whole, between the places deter- 
mined, after an interval of forty-five years, at Konigsberg and 
Yale, enhanced the significance of some minute discrepancies, 
the most considerable of which were held, with a strong show 
of reason, to imply that six of the objects on Bessel's list were 
only apparent members of the cluster.* They should indeed 
be regarded as pseudo-Pleiades, intruders into a company from 
which they will eventually be expelled through the irresistible 
effects of incompatible movements. Exempt from the influence 
of the current bearing Alcyone and its true associates slowly 
towards the south-south-east, they remain almost absolutely 
stationary in the background of the sky, and are accordingly 
m course of being left behind. 

The proper motion of Alcyone reverses, with approximate 
accuracy, the direction of the sun's progress through space. 
It may hence be regarded as parallactic, that is, transferred 
in appearance from our own. But the six stars lying apart 
from this perspective drift must (unless on the improbable sup- 
position of their being our fellow-travellers) be so remote, that 
the path traversed by the solar system in forty-five years, 
estimated to be at least twenty thousand millions of miles 

> Annates de VOhservatovre, t. xiv. ii. ; Contptes RenduSf t. Ixzxi. p. 6. 
» Monthly Notices^ vol. xliv. p. 357. 
» Trans. Yale College Observatory, vol. i. pt. i. 1887. 
* Dr. Elkin expressed this view with considerable reserve as regards four of 
the six stars. 


in length, dwindles as viewed from them — certainly with 
some help from foreshortening — to evanescence ! The smi 
seems stationary to them, as they seem stationary to us. 
Nor is their brightness such as to discredit the inference 
of their remoteness; they range in magnitude from 7'9 
to 9-2. 

Besides these half-dozen destined deserters, two stars 
(8 and 25 Pleiadum) , may, for an opposite reason, be presumed to 
stand aloof from the collection. Instead of lagging behind, they 
are hurrying on in front. They exhibit in excess the move- 
ment shared by the great majority of their seeming com- 
panions. They are then nearer to the sun, by perhaps one- 
third, than the veritable Pleiades, from which they, as well as 
the six ' fixed ' objects visually intermixed with them, should 
henceforth be carefully distinguished. 

The distance of Alcyone from the earth has never been 
measiured ; but it can be calculated, given the direction and 
velocity of the sun's translation, on the hypothesis that the 
proper motion of the star is simply a perspective effect of that 
translation. Now we know that the sun is travelling towards 
a point in the constellation Hercules not far from the star X ; 
and the rate of its journey is unlikely to fall short of fifteen, 
or to exceed twenty miles a second. Adopting the lower 
estimate, we find the distance of the Pleiades to be nearly 1,500 
billions of miles, or 250 light-years (parallax = 0"-013). And 
this may be considered a minimum value. 

Our own sun, thus prodigiously remote, would shrink to a 
star below the tenth magnitude, and much fainter, accord- 
ingly, than any of those measured by Dr. Elkin. There can 
be little doubt, in fact, that the solar brilliancy is surpassed 
by sixty to seventy of the Pleiades. And it must be, in some 
cases, enormously surpassed ; by Alcyone 1,000, by Electra 
480, by Maia nearly 400 times. Sirius itself takes a sub- 
ordinate rank when compared with the five most brilliant 
members of a group, the real magnificence of which we can 
thus in some degree apprehend. 

The scale of its construction is no less imposing. No 
udgment can of course be formed as to the interval of space 


(Separating any two of the stars belonging to it. All of them 
are seen projected indiscriminately upon the same plane, 
without regard to the directions in which they he one from 
the other. The line joining Maia, for instance, with Alcyone, 
may be foreshortened to any extent, or not at all. No 
criterion is at hand which we can apply. Of the dimensions, 
however, of the cluster as a whole, some notion can be 
gathered. For its shape — irrespective of some outlying 
streams of small stars — may be taken to be rudely globular ; 
and since a circle described from Alcyone as a centre with a 
radius of 48' includes all the principal stars, sixty of Elkin's 
sixty-nine, fifty-two of Bessel's fifty-three falling within it, 
the apparent diameter of the denser part of the aggregation 
cannot differ much from 96'. But the proportion of the radius 
to the distance ofa globe of known angular dimensions is easily 
arrived at, and here comes out (in round numbers) as one to 
seventy-one ; so that the bodies situated close to its surface 
are seventy-one times nearer to their central luminary than 
their central luminary is to us. If they revolve round it, it 
is at. the stupendous interval of (at least) twenty-one billion 
miles, costing light three and a half years to cross ; and the 
period of their circulation may well be reckoned by millions 
of years. Upon these dependent orbs, Alcyone shines with 
eighty-three times the lustre of Sirius in terrestrial skies; 
yet the presence of 67,000 Alcyones would only just com- 
pensate for the withdrawal of even such a diminished 
sun as brightens the firmament of Neptune. From stars 
more centrally placed, the chief of the cluster doubtless 
appears a veritable sun, although it may not be to all the 
primary light-giver. An assemblage hke the Pleiades dis- 
tributed round our sun would extend compactly three-quarters 
of the way to a Centauri, its feelers and appendages in- 
definitely farther. Hence there would be ample room in it 
/or secondary systems and particular associations of luminous 
bodies. And, in point of fact, the actual cluster contains 
several of Burnham's close double stars, presumably in 
mutual revolution, to say nothing of the doubtful companion 
of Atlas, which, distinctly visible only once to Struve in 1827, 

Q 2 


gave nevertheless some sign of its presence daring an occnlta- 
tion by the moon, January 6, 1876.^ 

All that is certain about the movements just beginning to 
be perceptible among the Pleiades, is that they are very small. 
Dr. Elkin concluded from the minuteness of the displacements 
brought to light by his measures, that * the hopes of obtaining 
any clue to the internal mechanism of the cluster seem not 
likely to be realised in an immediate future; ' and remarked 
the especial immobility of the brighter stars.^ Electra, alone 
of these, shifts as much as one second of arc in a century, 
whether along a curved or a straight line it is of course at 
present impossible to tell. The slightly swifter movements 
ascribed to a score of minor objects harmonise very imper- 
fectly (so far as can be judged) with preconceived notions of 
orderly circulation. They, on the contrary, rather suggest the 
advance of some slow process of systemic disintegration. M. 
Wolfs impression of centrifugal tendencies accordingly derives 
some confirmation from Dr. Elkin's chart, in which divergent 
movements seem to prevail, the region round Alcyone resem- 
bling, on a cursory view, a confused area of radiation for a 
flight of meteors, much more than the central seat of attraction 
of a revolving throng of suns. 

There are indications, however, that adjacent stars in the 
cluster do not pursue independent courses ; community of 
drift is perceptible in several quarters. At least four batches 
of stars seem to travel at nearly the same rate along ap- 
proximately parallel lines, and may perhaps eventually set up 
as independent systems. Thus, inquiries into the condition 
of the Pleiades afford as yet little countenance to the view 
that their dynamical relations are of a permanent kind. 
On this point differences of radial motion, such as the 
Potsdam spectrograph is capable of detecting, ought to prove 
highly instructive. We may learn from them, for instance, 
whether a uniform method of circulation prevails in the cluster. 
For if so, its constituent stars should be found approaching 
(comparatively to Alcyone) on the one side, receding on 
the other. The reality, too, of local groupings of apparent 

I Astr. NacK No. 2074. » Trans, Yale College Obs. vol. i. p. 101. 


movement might be similarly tested, if only the spectra of 
the objects concerned were bright enough for the purposes of 
photographic line-measurement. The Greenwich results show- 
ing Alcyone (17 Tauri) to be approaching the earth at the rate 
of about thirty-six miles a second, need confirmation all the 
more that since the journey of the solar system is directed 
obliquely ^om the Pleiades, it could have no share in pro- 
ducing a velocity necessarily shared by the entire physical 

A spectrum of the Sirian type characterises them. A 
simultaneous spectrographic impression obtained by Pro- 
fessor Pickering from close upon forty of these stars, January 
26, 1886, demonstrated the nearly identical quality of their 
light, and furnished ' strong confirmation of their common 
origin.' ^ Only in two cases a stronger ' K line ' recorded 
itself than such light ordinarily includes, and the divergence 
was, in one of the two, both accentuated and explained by 
diversity of motion. The star in question (s Pleiadum) has 
been already signalised as an incipient fugitive from the group 
to which it never truly appertained. 

The stars of the Pleiades, while shining with so poignant a 
lustre as to make the sky-background they are relieved upon 
show to the eye as blacker than elsewhere, are in reality wrapt 
and entangled in an immense cosmical cloud. Some indica- 
tions to this effect caught by optical means have been auto- 
graphically amplified to so surprising an extent that the dis- 
covery of the nebulous condition of the Pleiades ranks among 
the most important achievements of celestial photography. 

The * Merope nebula ' was compared by the late M. Tempel 
to a stain of breath upon a mirror. Discovered by him at 
Venice, October 19, 1869, it envelopes and stretches back in 
cometary shape from the star to which it is attached, cover- 
ing a space of about 35' by 20'.^ But this large size only 
makes its perception more difficult, by impairing the effect 
of contrast with the surrounding sky. With high magnifying 
powers (which imply narrow fields of view), it is on this 

* Mefnoir8 Amer. Acad, vol. xi. p. 215. 

> Astr, Naeh. No. 1290 ; Monthly Notices, toI. xl. p. 622. 


account completely invisible, and a haze so slight as to permit 
the observation of stars of thirteenth or fourteenth magnitude 
suffices to obliterate it. Thus, it has been only exceptionally 
seen, and has often been suspected to be variable. Tempel's 
contrary opinion, however, has of late been fully justified. 

The idea, too, was entertained both by Goldschmidt ^ and 
Wolf that the filmy veil flung round Merope was but a frag- 
ment of a larger whole, and, as time went on, glimpses were 
snatched of misty shreds and patches in connection with 
other members of the group. Alcyone appeared to Searle 
at Harvard College, November 21, 1875, as surrounded by 
whitish light ; ^ the attainment by the effusion about Merope 
of Electra and even Celseno was evident to Schiaparelli in 
1875,? and to Maxwell Hall in 1880 ; * while a remarkable view 
afforded to Mr. Common by his three-foot reflector, February 
8, 1880,* of three feebly luminous blotches between Merope 
and Alcyone, prompted his comment that ' there is a great 
deal yet to be settled as to the extent and number of the 
nebulsB in this cluster.' 

Its significance, however, became apparent only when 
photography was brought to bear upon the subject. The 
first nebula discovered by the new method was a small spiral 
appendage to the star Maia, which printed itself on plates 
exposed by the MM. Henry, each during three hours, in 
December 1885.® Only the accumulating faculty of the 
* chemical retina * could have revealed the presence of an 
object so excessively faint in a telescopic sense ; but what is 
known to exist is, by that alone, rendered more than half 
visible, and the Maia nebula was accordingly discerned, 
February 5, 1886, with the Pulkowa thirty-inch refractor, 
then newly erected, and later with smaller instruments.^ 

Besides the Maia vortex, the Paris photographs depicted 
a series of nebulous bars on either side of Merope (partly 

* Les MondfSt t. iii. p. 529. ' Harvard Annalst vol. xiii. p. 74. 
» Astr. Nach. No. 2045. 

* Monthly Notices, vol. xli. p. 315. » Ibid, vol. xl. p. 376. 

* Similarly recorded a month earlier at Harvard Collegei it was taken for a 
flaw in the negative. 

' Astr. Nach. Nob. 2719, 2736, 2730. 


seen by Mr. Common), and a curious streak extending like a 
finger-post from Electra towards Alcyone. But all these were 
mere samples of what lay behind. Impressions of the Pleiades 
secured by Mr. Eoberts with his twenty-inch reflector in 
October and December 1886, showed the whole western side 
of the group to be involved in one vast nebulous formation.* 
* Streamers and fleecy masses ' of cosmical fog seem, in these 
astonishing pictures, almost to All the spaces between the 
stars, as clouds choke a mountain valley. The chief points 
of its concentration are the four stars Alcyone> Merope, Maia, 
and Electra ; but it includes as well Gelaeno and Taygeta, and 
is traceable southward from Asterope over an arc of 1° 10'-. 
Tempers nebula appears in its proper elliptical form in ad- 
dition to the barred structure stamped upon the Paris plates ; 
the little curved train of Maia is visible, though immersed in a 
far wider luminosity, reaching nearly to Asterope; Electra 
projects towards Alcyone a dim shaft, to which a thin streamer, 
' resembling a detached nebulous straight line,' runs parallel 
further south.^ These photographs, in fine, as Mr. Wesley 
wrote, ' not only prove beyond a doubt the existence of the 
much-disputed Merope nebula, but they also combine and 
harmonise in a very satisfactory manner the apparently irre- 
concilable drawings.' * 

The matter was not, however, allowed to rest here. Early 
in 1888 the MM. Henry succeeded in giving to several plates 
exposures of four hours, with results identical in each case, and 
very curious. Their nature can be estimated from the frontis- 
piece, which reproduces the final chart of the Pleiades prepared 
by the MM. Henry. The greater part of the constellation 
is shown in it as veiled in nebulous matter of most unequal 
densities. In some places it lies in heavy folds and wreaths, in 
others it barely qualifies the darkness of the sky-ground. The 
details of its distribution come out with remarkable clearness, 
and are evidently to a large extent prescribed by the relative 
situations of the stars. Their lines of junction are frequently 
marked by nebulous rays, establishing between them, no 

> Monthly Notices^ vol. xlvii. p. 24. 
• Ibid, p. 90. » Jour, Liv. Attr, Soc, vol. v. p. 150. 


doubt, relations of great physical importance ; and masses of 
ilebola, in numerous instances, seem as if pulled out of shape 
and drawn into festoons by the attractions of neighbouring 
stars. But the strangest exemplification of this filamentous 
tendency is in a fine, thread-like process, 8'' or 4" wide, but 
85^ to 40' long, issuing in an easterly direction from the edge 
of the nebula about Maia, and stringing together seven stars 
met in its advance, ' like beads on a rosary/ ' The largest of 
these is apparently the occasion of a slight deviation from its 
otherwise rectilinear course. A second similar, but shorter 
streak runs, likewise east and west, through the midst of the 

Whether these luminous highways are due to material 
condensations, or merely indicate tracks of electrical excite- 
ment, they are equally significant upon one point. The con- 
nection by their means of rows of stars virtually demonstrates 
their real alignment, and thus considerably strengthens the 
presumption that the linear arrangement prevalent in clusters 
is no optical illusion, but depends upon intrinsic conditions, 
the outcome of universal laws. 

Investigations of the Pleiades have led to many surprises ; 
possibly the supply of them is not even yet exhausted. The 
main result so far has been to exhibit the group as com- 
bining with such singular completeness the properties of. a 
great nebula with those of a cluster, that we are inevitably 
led to regard the gap between these two kinds of aggregation 
as less wide than had heretofore been supposed. 

Stars of all orders are there gathered together into (it 
might be said) a miniature sidereal system. The largest are 
of such 'surpassing glory' as to dim by comparison the 
splendour of Sirius and Vega; the least are probably as 
inferior to them as the moons of Mars are to Jupiter. The 
'act of order' in this 'peopled kingdom' is not easy to 
divine, but the mutual relations of its denizens are at once 
perceived to be highly intricate. Within the wide frame- 
work of the association room is found for subordinate 
groupings of various characters and degrees of closeness, 
* Mouchez, Comptes Rendus^ t. cvi. p. 912. 


from stars far apart, but * drifting ' in company, to pairs as 
unmistakably united by contiguity as two nuts within the 
same shell. Thus, the polity governing the entire system 
of the Pleiades would seem to be of the federative kind. Nor 
can we be yet sure that its bonds, while evidently so loose as 
to give unshackled play to local liberties, are nevertheless 
sufficiently strong to restrain the slow workings of disruptive 



About five hundred qlusters are at present tolerably well 
known to astronomers, and a large number besides, their 
character rendered ambiguous by distance, are probably 
included among both ' resolvable ' and * unresolved * nebulae. 
Such aggregations may be broadly divided into 'irregular* 
and ' globular ' clusters. Although, as might have been ex- 
pected, the line of demarcation between the two classes is by 
no means sharply drawn, each has its own marked pecu- 

Irregular clusters are framed on no very obvious plan; 
they are not centrally condensed, they are of all shapes, and 
their leading stars rarely occupy critical positions. The stars 
in them are collected together, to a superficial glance, much 
after the fashion of a flock of birds. Alcyone, it is true, 
seems of primary dignity among the Pleiades, and the Pleiades 
may be regarded as typical of irregular clusters; yet the 
dominance, even here, of a central star may be more apparent 
than real. 

The arrangement of stars in clusters is, nevertheless, far 
from being unmethodical, even though the method discernible 
in it be not of the sort that might have been antici- 
pated. It seems, indeed, inconsistent with movements in 
closed curves, and suggests rather the description of hyperbolic 
orbits. Obviously, however, its true nature must be greatly 
obscured to our perception by the annulment, through per- 
spective, of the third dimension of space, whereby independent 
groupings, flattened down side by side, are rendered scarcely, 

* See Nature, vols, zzzyili. p. 365, xxxix. p. 13. 


if at all, distinguishable* That they should under these cir- 
cumstances be to any extent traceable is more surprising 
than that they should sometimes be inextricably entangled 
■with sprinkled stars belonging to the fore- or background. 
The nebulous linking together of a septuple set in the Pleiades 
assures us, however, that they are traceable, and that star- 
alineations are not illusory. 

Nearly all observers have been impressed with the streaming 
and reticulated structure characterising many stellar assem- 
blages. Thus where the feet of the Twins dip into the 
Milky Way, an object is encountered so 'marvellously striking ' 
with a large telescope, that ' no one could see it for the first 
time,' Mr. Lassell declared, * without an exclamation.' A field 
19' in diameter ' is perfectly full of brilliant stars, unusually 
equal in magnitude and distribution over the whole area. 
Nothing but a sight of the object itself can convey an idea of 
its exquisite beauty.'* Admiral Smyth described it as 'a 
gorgeous field of stars from the ninth to the sixteenth magni- 
tudes, but with the centre of the mass less rich than the rest. 
From the small stars being inclined to form curves of three 
or four, and often with a large one at the root of the curve, 
it somewhat reminds one of the bursting of a sky-rocket.' * 
A beautiful photograph of this cluster,^ taken by the MM. 
Henry, March 10, 1886, exhibits not less than two thousand 
of its components disposed in a kind of starfish pattern, the 
branches often connected by drooping chains, and composed 
in detail of sinuous lines, or the * fantastically crossing arcs ' 
of stars noticed by Father Secchi,* 

The * wonderful looped and curved lines ' of conformation 
visible in a cluster in Auriga (M 87) attracted the attention 
both of D'Arrest and Lord Eosse ; ^ about one hundred con- 
nected stars in Ophiuchus (M 28) ' run in lines and arches ; ' * 
a coUection in Sagittarius (N. G. C, 6416) makes a ' zigzag ' 

> Monthly Notices, vol. xiv. p. 76. « Cycle, p. 168 (Chambera's ed.). 

* M 35 B N. G. C. 2168. Nebnlss and clusters are throughout this volume 
distinguished, when among the 103 catalogued by him, by Messier's well- 
known numbers, otherwise by Dreyer's in the New General Catalogue, 

♦ Atti deW Accad, Pont. t. vii. p. 72. 

» Trans. R. Irish Acad, vol. ii. p. 51. • Phil. Trans, vol. cxxiii. p. 460. 


figure. The constituents of a large group near the Poop 
of Argo (N. G. C, 2567) struck the elder Herschel by an 
arrangement * chiefly in rows,' illustrative, to his mind, of 
the mechanical complexities of such systems. Each row, he 
observed, while possessing its own centre of attraction, will 
at the same time attract all the others ; nay, ' there must be 
somewhere in all the rows together the seat of a preponde- 
rating clustering power which will act upon all the stars in the 
neighbourhood.* ' Speculations, indeed, upon the dynamical 
relations of ' stars in rows ' are still premature, nor are they 
likely^ for some time to come, to be accounted as *of the 
order of the day.' But the continual recurrence in the 
heavens of this mode of stellar aggregation cannot fail to 
suggest the development of plans of systemic dissolution 
and recomposition on a grand scale, and involving the play 
of, to us, unknown forces. 

The more attentively clusters are studied, the more in- 
tricate their construction appears. That which challenged 
Herschel's notice is not si;ngular in intimating a league of 
several co-ordinate groups. There is rarely evidence in the 
conformation of irregular clusters of their being governed 
from a single focus of attraction ; there are frequent indica- 
tions of the simultaneous ascendency of several. A cluster in 
Sagittarius (N. G. C, 6451) is distinctly bifid. It was re- 
marked by Sir John Herschel at Feldhausen as ' divided by a 
broad, vacant, straight band ' ; ^ and his figure shows the 
separation as absolutely complete, the sections. 

Like cliffs which had been rent asunder, 

facing each other with a chasm between. 

A splendid cluster in Sobieski's Shield (M 11) seems to be 
essentially trifid. Sir John Herschel, indeed, succeeded, by 
the use of high powers, in breaking it up * into five or six 
distinct groups with rifts or cracks between them.'* But 
Father Secchi perceived in it a three-lobed central vacuity ; * 
and the photograph reproduced in fig. 84 exhibits two great 

> Pha. Trans, vol. civ. p. 269. • Cape Observations, p. 116. 

• Pha. Trans, vol. cxxiii. p. 462. « AtU delV Accad. Pont t. vii. p. 75. 


wings of stars in its neighbourhood to about the fourteenth 
magnitude, with a connecting mass between. Some partial 
clearings, however, accentuating the structure of the star- 
clouds, have been to a great extent obliterated in the process 
of reproduction. The cluster itself, compared by Admiral 
Smyth to a ' flight of wild ducks,' forms a sort of nucleus to 
the entire. The original of the picture was taken by Mr. E. E. 

Fm. 84.— Star Cluster in Sobieski's Shield (M 11) photographed by 
Mr. £. £. Barnard at the Lick Observatory. 

Barnard, of the Lick Observatory, with a portrait-lens of just 
six inches aperture and thirty-one focus, the exposure lasting 
two hours and three-quarters. The diameter of the field is 
5-7 degrees. 

This * glorious object * (as Sir John Herschel called it) 
can just be made out with the naked eye on a perfectly clear 
night. Halley mentioned it in 1716 as * of itself but a small 
obscure spot, but with a star that shines through it which 
makes it the more luminous.' ^ Some years later, Derham 
found it to be ' not a nebulose, but a cluster of stars, some- 

» Phil, Trans, vol. xxix. p. 392. 


what like that which is in the Milky Way/ ' A catalogue of 
two hondred of the components (several of which apppear to 
be variable) prepared in 1870 by M, Helmert of the Hamburg 
Observatory,* provides material for the future investigation of 
relative changes. 

The presence in a cluster in Monoceros (N. G. C, 2269) of 

* a double seat of preponderating attractions ' was observed by 
Sir William Herschel ;• and a throng of some two hundred 
stars in Cancer (M 67), discernible with an opera-glass, falls 
no less obviously into two divisions/ In a collection seen at 
Parsonstown to be riddled with absolutely dark ' lanes and 
openings* (N. G. C,, 2648), the principle of * local self-govern- 
ment' has evidently been already carried a long way. A 

* reticulated mass of small stars ' in Gygnus (N. G. C, 6819) 
was there described as ' a most gorgeous cluster, fuU of holes ; ' 
and the drawing published by Lord Bosse depicts a winding 
ribbon of stars inclosing three blank circular spaces of sym- 
metrically varying diameters. 

Among the ' curiosities ' of the heavens are to be reckoned 
clusters within clusters. In one such instance, a large loose 
collection in Gemini (N. G. C, 2381) involves a neat group of 
' six or seven stars close together, and well isolated from the 
rest.' ® Another (N. G. C, 2194) occurs where the Milky Way 
passes between Gemini and Orion. 

Star-groupings of singularly definite forms are often met 
with. A triangular swarm (N. G. C, 7826) presents itself in 
Cetus ; a rectangular area in Vulpecula (N. G. C, 6802) is 
densely strewn with fine star-dust. Clusters shaped Uke 
half-open fans are tolerably numerous. One situated in 
Gemini, if removed to a sufficient distance, would appear, 
according to Sir John Herschel, *as a fan-shaped nebula 
with a bright point like a star at the vertex.' Another 
specimen of an * acutangular ' cluster 2' in length (N. G. C, 

• Phil. Trans, vol. xxxviii. p. 72. 

■ Publicationen der Hamburger Stemwarie% No. 1, 1874. 
» Phil. Trans, vol. civ. p. 268. 

* Smyth, Cycle, p. 241 ; L. Fenet, L^Astronomis, t. vi. p. 146. 
» Trans. B. Dublin Soc. vol. ii. p. 66. • Ibid. p. 66. 


7510), is bounded by ' two principal lines of stars drawing to 
one.' * 

Bed or double stars frequently appear in some sort to 
dominate stellar assemblages. The chances of optical juxta- 
position must indeed produce compound objects most freely 
where stars are most crowded, yet when they are marked 
out (as occasionally happens) both by superiority of lustre 
and by distinction of place, some significance may be attached 
to their presence. Thus, each of the oblique arms of a * cruci- 
form ' group in Auriga (M 88), photographed at Paris, January 
28, 1887, carries a pair of conjoined stars brighter than the 
rest.^ A * superb cluster ' in Monoceros (N, G. C, 2548), 
standing out from a background of sky * singularly dotted 
over with infinitely minute points,' has a double star in its 
most compressed part.* The central star in Prcesepe is double, 
and there are many examples of more restricted gatherings 
round a composite luminary. 

Groups apparently ruled by a conspicuous coloured object 
are met with in Auriga (N. G. C, 1857) and Cygnus (N. G. C, 
7086). In Cygnus, too, is an oval annulus, 4! across (N. G. C, 
7128), of stars centrally surrounding a ruddy one of the ninth 
magnitude. A similar elliptical group, with a double substi- 
tuted for the red star, constitutes a quasi-nucleus for one of 
the two great adjacent clusters in Perseus (N. G. C, 869). 
This superb object, like the still richer group (N. G. C, 884) it 
immediately precedes, was regarded by Herschel as merely a 
protuberance of the Milky Way, and its galactic affinities are 
undoubtedly very close. The two together form a telescopic 
pageant such as, in the wildest flights of imagination, 
Hipparchus could little have dreamed would one day be un- 
rolled before the eyes of men out of the * cloudy spot ' in the 
sword-handle of Perseus which he (it is said) was the first to 
detect. Although the outlines of the two clusters can be 
brought within the same field of view, they are held to be 
really disconnected ; * it is scarcely probable, however, that 
they were originally so. The second, and more considerable 

> PfciZ. Tram, vol. cxxiii. pp. 476, 603. « Smyth, CycU^ p. 140, 

• J. Herschel, Vhih Trans, vol. oxxiii. p. 386. * Smyth, Cycle, p. 60. 


(known as * x P^i^sei '), was micrometrically InveBtigated by 
Vogel in 1867-70, photographically by 0. Lohse in 1884,' 
with the result, from the comparison of 172 stars, of demon- 
strating their fixity during an interval certainly too short for 
the development into visibility of such tardy movements as 
were alone likely to be in progress. A rapid spectroscopic 
survey executed by Vogel with the Berlin nine-inch refractor, 
March 80, 1876,* disclosed no peculiarity in the light of any 
of the collected stars, although some of them have been called 
red, ' pale garnet,' and even * ruby.* Their brilliancy suggests 
that this magnificent assemblage may be less exorbitantly 
distant from the earth than most other objects of its class. 
A fine photograph of the * double cluster,' in which the 
'festoon-like groupings' composing it are conspicuous, was 
taken by Mr. Boberts at MaghuU January 18, 1890.' 

The famous tintf^d cluster about /c Crucis can only be seen 
from southern latitudes. And it must be confessed that, with 
moderate telescopic apertures, it fails to realise the effect of 
colour implied by 8ir John Herschers comparison of it to a 
* gorgeous piece of fancy jewellery.' A few reddish stars 
catch the eye at once; but the blues, greens, and yellows 
belonging to their companions are pale tints, more than half 
drowned in white light. Some of these stars are suspected of 
considerable mobihty. During his visit to the Cape, Herschel 
determined the places of 110, all included in an area of about 
^ of a square degree,^ and the process was repeated and ex- 
tended to 180 components by Mr. H. C. Bussell of Sydney in 
1872.* The upshot was to bring out discrepancies which, if 
due to real movements, would be of extreme interest. But 
Herschel's measurements were necessarily too hasty to be 
minutely reliable, and changes depending upon their authority 
need to be confirmed by continuance before they can be un- 
reservedly accepted. The same qualification appUes to M. 
Cruls's supposed discovery of orbital revolution in three 
double stars within the precincts of the cluster.* 

In the constellation Cancer may be seen any fine night in 

• Astr. Nach, No. 2660. « Der Stemhaufen % Persei, p. 81. 

• Monthly Notices^ vol. 1. p. 316. * Cape ObaervatumSf p. 17. 

» Monthly Notices, vol. zzziii p. 66. * Comptea Rendus, t. Ixxzix. p. 435. 


;.' ' •■ i:^ ^: 

M.j. . •• '.'. ! 


winter, a blot of dim light placed midway between two fourth- 
magnitude stars. The stars were called by the ancients the 
Asses, Astdli, the interposed cloudlet representing to their fancy 
a * Manger,' Prcesepe. Since its disappearance was reckoned a 
sure presage of rain,^ a good deal of popular attention was 
paid to it, and its stellar constitution was one of the earliest 
telescopic discoveries ; but only prelimuiary steps have been 
taken towards its exact investigation. Of its components, 
thirty are measurable on Rutherfurd's photogiaphs, and 363 
were mapped over an area of three square degrees by C. Wolf 
some sixteen years later, eighty-two among them being care- 
fully determined as ' fundamental stars ' for the rest, by the 
same methods used for the Pleiades.^ As yet, however, no 
materials are ripe for comparison. The date (1870) of Pro- 
fessor Asaph Hairs catalogue of 151 stars is too recent 
for the purpose; the results of Winnecke's observations of 
Prsesepe with the Bonn heliometer have not been made 
pubUc. Hence we are uninformed as to the nature of what- 
ever changes may be progressing in this system. 

The particles of a drop of water are not in more obvious 
mutual dependence than the constituent stars of globular 
clusters ; ' the most magnificent objects,' in the elder Herschers 
opinion, ' that can be seen in the heavens.' Were there only 
one such collection, the probability of its separate organisa- 
tion might be reckoned * infinitely infinite,' and one hundred and 
eleven of them were enumerated by Sir John Herschel in 1864. 
It does not, however, follow that the systems thus constituted 
are of a permanent or stable character ; their configuration, 
in fact, points to an opposite conclusion. There may, of 
course, be an indefinite number of arrangements by which the 
dynamical equilibrium of a * ball of stars * could be secured ; 
there is only one which the present resources of analysis en- 
able us distinctly to conceive. This was adverted to, many 
years since, by Sir John Herschel. Equal revolving masses, 
uniformly distributed throughout a spherical space, would, he 

• Aratu8, Diosemeia, vv. 160-180, 265 ; Theophrastus, De Signis Pluviarunif 
ed. HeinsiuB, p. 419. 

" ContpU's Ucndus^ t. xcv. p. 333. 



showed, be acted upon by a force varying directly as the dis- 
tance from the centre. The reason of this is easily seen ; for 
the further out a component of such a system is located, the 
more matter there will be inside, and the less outside its orbit. 
The strength of the central pull thus reaches a maximum at 
the surface of the sphere, the velocity by which it is balanced 
growing in the same proportion. Ellipses described under these 
conditions would all, accordingly, have an identical period ; 
whatever their eccentricities, in whatever planes they lay, in 
whatever direction they were traversed, each would remain 
invariable ; and the harmony of a system in which no pertur- 
bations could possibly arise would remain unbroken for ever, 
provided only that the size of the circulating bodies, and the 
range of their immediate and intense attractions, were in- 
significant compared with the spatial intervals separating 

But this state of nice adjustment is a mere theoretical 
possibility. There is no likelihood that it has anywhere an 
actual existence ; and the stipulations, upon compliance with 
which its realisation strictly depends, are certainly disre- 
garded in all the stellar groups with which we have any close 
acquaintance. The components of these are neither equal, 
nor equably distributed. Central compression, over and above 
the merely apparent effect of the gradually increasing depth 
of the star-strata presented to the eye, is the rule in globular 
clusters. Three distinct stages of condensation w^ere noted 
by Herschel in the efifulgent southern object called * 47 Tou- 
cani ; ' real crowding perhaps intensifies the * blaze ' in the 
middle of the superb group between 17 and f HercuKs (M 13 ; 
see Plate II.) ; in other cases, the presence of what might be 
termed a nuclear mass of stars is evident. Here, then, the 
*law of inverse squares' must enter into competition with 
the * direct' law of attraction, producing results of extra- 
ordinary intricacy, and giving rise to problems in celestial 
mechanics with which no calculus yet invented can pretend 
to grapple. 

Sir John Herschel allowed the extreme diflBculty of even 
* Outlines of Astronomy^ 9th ed. p. 636. 


imagining the * conditions of conservation of such a system 
as that of to Centauri, or 47 Toucani, &c., without admitting 
repulsive forces on the one hand, or an interposed medium 
on the other, to keep the stars asunder/ * Compacted into 
a whole, they might, he thought, instead of revolving in- 
dividually, be supposed to rotate in their corporate capacity 
as a single body. But the establishment in such aggrega- 
tions of a * statical equilibrium ' by means of an ' interposed 
medium' is assuredly chimerical. The hypothesis of their 
rotation as one mass is countenanced by no circumstance 
connected with them, and is decisively negatived by their 
irregularities of figure. The sharp contours of bodies whirl- 
ing on an axis are nowhere to be found among these objects. 
Their streaming edges betray a totally different mode of 

Globular clusters commonly present a radiated appearance 
in their exterior parts. They seem to throw abroad feelers 
into space. The great cluster in Hercules is not singular 
in the display of ' hairy-looking, curvilinear ' branches. That 
in Canes Venatici (M 3) has * rays running out on every side ' 
from a central mass, in which * several small dark holes ' were 
disclosed by Lord Rosse's powerful reflectors ; * showing pretty 
plainly that the spiral tendency, visible in the outer regions, 
penetrates in reality to the very heart of the system. From 
a well-known cluster in Aquarius (M2), * streams of stars 
branch out, taking the direction of tangents.' ^ That in 
Ophiuchus (M 12) is provided with long straggling tentacles, 
of a * slightly spiral arrangement,' according to the late Lord 
Eosse. And a remarkable assemblage in Coma Berenices 
(M 53) was described by Herschel and Baily as ' a fine com- 
pressed cluster with curved appendages like the short claws 
of a crab running out from the main body.' ^ The peculiarity 
in question is the more significant that it is shared ly many 
undoubted nebulae. 

We find it difficult to conceive the existence of * streams 
of stars ' that are not Jioicing ; and accordingly the persistent 

• Cape Observatio7is, p. 139. ' Trans. R. Dub, Soc. vol. ii. p. 132. 

• Ibid. p. 102. * Phil. Trans, vol. cxxiii. p. 45S. 

u 2 


radial alignment of the components of clusters inevitably 
suggests the advance of change, whether in the direction of 
concentration or of diffusion. Either the tide of movement 
is setting inward, and the * clustering power ' (to use a favour- 
ite phrase of Sir William Herschel's) is still exerting itself to 
collect stars from surrounding space; or else a centrifugal 
impulse predominates, by which full-grown orbs are driven 
from the nursery of suns in which they were reared, to seek 
their separate fortunes and enter on an independent career 
elsewhere. The somewhat hazardous conjecture that the 
process of their development may be attended by an increase 
of tangential velocity, appears to receive some countenance 
from the general superiority in brightness of outlying stars 
over those more centrally situated in globular clusters. But, 
in truth, the question as to whether separatist or aggrega- 
tionist tendencies prevail in them is of too recondite a nature 
for profitable discussion at present. All that can be said is 
that, after the lapse of some centuries, photographic measure- 
ments may help towards deciding it. Even then, however, 
the answer may long appear dubious. For it is not im- 
possible (though far from probable) that processions of stars, 
compelled by the attractive power of the cluster to deviate 
from their march past it, may here be arriving, there depart- 
ing, after having effected an hyperbolic sweep round its denser 
portion. Comets, doubtless, from time to time dash through 
the system ; it cannot be affirmed offhand that stars may not 
do likewise. 

An object visually resembling a blurred star below the 
fourth magnitude was named by Bayer ck> Centauri. It never 
rises in these latitudes, but Herschel's great reflector revealed 
it to him at the Cape as a 'noble globular cluster, beyond all 
comparison the richest and largest object of the kind in the 
heavens.' ^ The stars contained in it are literally innumer- 
able; they are all excessively minute, and approximately 
equal, though irregularly distributed into little knots and 

The loveliness of the cluster 47 Toucani near the Lesser 

• Cape Observations^ p. 21. 


Magellanic Cloud was, to Herschel'sview, set off by a diversity 
of colour between an interior mass of rose-tinted stars and 
marginal strata of purely white ones.^ This feature, however, 
has met with no later recognition ; and to the present writer 
in 1888, the object appeared of the same silvery sheen through- 
out. Its diameter, inclusive of stragglers giving it the usual 
radiated aspect, measures about 20'; apart from them, it covers 
an area roughly equal to one quarter of the lunar disc, and 
is so obvious to unaided sight, that for several nights after his 
arrival in Peru it was mistaken by Humboldt for a comet.^ 
About midway between the centre and circumference of 
47 Toucani, a pair of stars much brighter than the rest are 
placed, perhaps by casual projection, although the blankness 
of the surrounding sky diminishes the probability of this 
being the case. 

The gradations of lustre among the constituents of such 
assemblages commonly range over three or four magnitudes. 
Nor by any means unmethodically. As a general, if not an 
invariable rule, the smaller stars are gathered together in the 
middle, while the bright ones surround and overlay them on 
every side. Thus, of a magnificent cluster in Sagittarius 
(M 22), known since 1665, the central portion accumulates the 
light of multitudes of excessively minute, and is freely sprinkled 
over with larger stars. The effect, which probably corre- 
sponds with the reality, is as if a globe of stars of fifteenth 
were enclosed in a shell of stars of eleventh magnitude, some 
of these being naturally projected upon the central aggregation. 
Sir John Herschel remarked of a cluster in the southern con- 
stellation of the Altar (N. G. C, 6752) : * The stars are of two 
magnitudes ; the larger run out in lines like crooked radii, the 
smaller are massed together in and around the middle.' ^ A 
similar structure was noted by Webb * in the Canes Venatici 
and Coma Berenices clusters already mentioned (M 3, M 53), 
as well as in an imposing collection in Libra (M 5), discovered 
by Kirch in 1702, the more condensed part of which (compared 

* Cape Observations, p. 18. * Cosmos (Ott6's trans.), vol. iii. p. 1( 2 

" Cajye Observation St p. 11^. * Student^ vol. i. p. 160. 


by Sir John Herschel to a snow-ball), seems as if ' projected on 
a loose, irregular ground of stars.' * 

Irregularities of distribution in clusters assume at times 
a highly enigmatical form. At Parsonstown, in 1850,* three 
* dark lanes,' meeting at a point considerably removed from 
the centre, were perceived to interrupt the brilliancy of the 
globe of stars in Hercules (M 13). They were afterwards 
recognised by Buffham and Webb, and recorded themselves 
with emphasis in a photograph by Mr. Eoberts, of which 
Plate II. is a reproduction. On the original negative, one of 
the * lanes ' can clearly be seen to lead to a small but perfectly 
symmetrical oval ring of stars, surrounding one centrally 
situated ; while further out, partial, clearings recur in the same 
shape as before, markedly enough to demonstrate the dominat- 
ing influence of some kind of law in their production. Globular 
clusters in Ophiuchus (M 12), in Pegasus (M 15),^ and in 
Canes Venatici (M 3), appear to be similarly tunnelled . Pre- 
conceived ideas as to the mechanism of celestial systems are 
utterly confounded by phenomena not easily reconcilable with 
the prosecution of any orderly scheme of circulatory move- 
ment. Even if absolutely denuded of stars, the extensive 
vacancies indicated by the visibility of dusky rifts must be in 
part obliterated by the interposed light of the surrounding 
star-layers. They can hence become perceptible only when 
most pronounced ; and are likely to exist, less fully developed, 
in numberless cases where they defy detection. 

Differences of distance are alone adequate to account for 
the variety of texture observable in globular clusters. ^1^^* 
in Aquarius, for instance, likened by Sir John Herschel to * a 
heap of golden sand,' might very well be the somewhat coarse- 
grained Hercules group withdrawn as far agam into space. 
At a still further stage of remoteness, the appearance w^ould 
presumably be reached of a stellar throng in the Dolphin 
(N. G. C, 6934), which, with low powers, might pass for a 
planetary nebula, but under stronger optical compulsion 
assumes the granulated aspect of a true cluster. And many 

' PJiil Trans, vol. cxxiii. p. 359. « j^/j. yoj. dj. p. 732. 

» Webb, Ccl. Objects, p. 372. 

Plate 2. 


Photographed by Isaac Roberts, F.R.A.S.^ May azW, 1887. Exposure one hour. 


such, their genuine nature rendered impenetrable by excessive 
distance, are doubtless reduced to the featureless semblance of 

* irresolvable 'nebulee, 

But there are real, as well as apparent, diversities in these 
objects. Although smaller and more compact clusters must, 
on the whole, be more remote than large, loosely-formed 
ones, yet *this argument,' Sir William Herschel remarked, 

* does not extend so far as to exclude a real difference which 
there may be in different clusters, not only in the size, but 
also in the number and arrangement of the stars/ There 
may be globular clusters with components of the actual 
magnitude of Sirius ; others, optically indistinguishable from 
them, may be aggregated out of self-luminous bodies no larger 
than Mars, or even than Ceres, or Pallas. Our total ignorance 
of the reallocation in space of such objects, leaves us without 
the means of judging. Nor is ignorance likely to be replaced 
by knowledge for an indefinite time to come. We can, how- 
ever, suppose it to be so replaced in a particular instance, and 
trace the consequences. Let our example be the great cluster 
in Hercules. 

* This is but a Uttle patch,' Halley wrote in 1716, * but it 
shows itself to the naked eye when the sky is serene, and the 
moon absent.' ^ Messier termed it ' nebuleuse sans etoiles ' ; * 
yet a * twinkling ' indicative of its stellar character may be 
caught with a telescope four inches in aperture ; and a power- 
ful instrument resolves it to the core. Within the precincts 
of Halley's 'little patch,' Sir WiUiam Herschel estimated 
fourteen thousand stars to be 'cribb'd, cabined, and confined ' ! 

The apparent diameter of this object, including most of 
the ' scattered stars in streaky masses and lines,'^ which form 
a sort of * glory ' round it, is 8' ; that of its truly spherical 
portion may be put at 5'. Now, a globe subtending an angle 
of 5' must have (because the sine of that angle is to radius 
nearly as 1 : 687) a real diameter -^^ of its distance from the 
eye, which if we assume to be such as would correspond to a 
parallax of yV ^^ ^ second, we find that the cluster, outliers 

> Phil. Trans, vol. xxix. p. 392. « Conn, des Temps, 1784, p. 233. 

" Phil. Trans, vol. cxxiii. p. 458. 


apart, measures 558,000 millions of miles acrose. Light, in 
other words, occupies about thirty-six days in traversing it, 
but sixty-five years in journeying thence hither. Its compo- 
nents may be regarded, on an average, as of the twelfth magni- 
tude ; for although the divergent stars rank much higher in the 
scale of brightness, the central ones, there is reason to believe, 
are notably fainter. The sum total of their light, if concen- 
trated into one stellar point, would at any rate very little (if 
at all) exceed that of a third-magnitude star. And one star 
of the third is equivalent to just four thousand stars of the 
twelfth magnitude. Hence we arrive at the conclusion that 
the stars in the Hercules cluster number much more nearly 
four, than fourteen thousand. 

If, then, four thousand stars be supposed uniformly distri- 
buted through a sphere 658,000 miUion miles in diameter, an 
interval of 28,800 million, or more than ten times the distance 
of Neptune fnom the sun, separates each from its nearest 
neighbour.^ Upon a spectator in an intermediate situation, eix 
stars (besides crowds of gradated inferiority) would then blaze 
with about 3,300 times the lustre that Sirius displays to us. 
Yet since a million and a half of stars of even this amazing 
brilliancy would be needed to supply the light we receive from 
the sun, the general illumination of the cluster can only 
amount to a soft twiUght, excluding, it is true, the possibility 
of real night on any globe placed near its centre. 

At its surmised distance, our sun would appear as a star 
of 7*5 magnitude ; it would shine, that is to say, about sixty- 
three times as brightly as an average one of the grouped 
objects. Each of these, accordingly, emits -^^^ of the solar 
light ; and if of the same luminosity relative to mass, as the 
sun, it exercises just -^^ of the solar attractive power. The 
mass of the entire system of four thousand such bodies is 
thus equal to that of eight suns. This, however, may be 
regarded as a minimum estimate. The probabilities are in 
favour of the cluster being vastly more remote than we have 
here assumed it to be ; hence composed of larger or brighter, 

* See Mr. J. E. Gore*8 similar calculation froui different data, in Jour. Liv. 
Astr. Soc. vol. V. p. 1G9. 



and presumably more massive, individual bodies than results 
from our calculation. 

The relations of clusters and nebulae are evidently very 
close; but they are only just beginning to be effectually 
studied. The conjecture is, however, already fully justified 
that the two classes of object form an unbroken series — that 
clusters exist in every stage of development from nebulae, and 
that the advancing condensation of many nebulsB will eventu- 
ally transform them into veritable clusters. Suggestions to 
this effect derived from analogies of form * are corroborated 
by numerous recent observations of the actual co-existence 
with grouped stars of nebulous masses. The Pleiades is not 
the only example of a hybrid system. A beautiful, bright 
cluster of the same general character in Sagittarius (N. G. C, 
6530) is obviously connected with a great nebula (M 8), in 
the meshes of which it seems as if entangled. The photo- 
graphic investigation of this mixed display may be expected 
to yield results of high interest. 

That potent method will also, it is hoped, be applied with- 
out delay to a combination of stars and nebulae without a 
known parallel in the heavens. Mr. E. E. Barnard discovered 
at Lick in 1889 a cluster in Monoceros (N. G. C, 2244) to be 
completely surrounded by a vast nebulous ring 40' in diameter.'* 
The interior appeared perfectly free from luminous, haze, the 
stars shining on an absolutely black sky. 

Two rich clusters have long been known to include each a 
nebula of the planetary kind. One in Argo (N. G. C, 2818) 
has a central vacuity conspicuously occupied by a nebulous 
disc 40" across; the other (M 46), in the same constellation, 
displays within its borders a fine annular nebula.^ It is 
difficult, if not impossible, to believe either projected casually 
into such a remarkable position. Mr. Eoberts's photograph, 
too, of the Hercules cluster at least suggests the presence of 

* Lockyer, Proc. R. Society^ vol. xliv. p. 29. 

' Astr. Nach. No. 2918 ; Swift, Sidereal Mess. vol. ix. p. 47 ; Backhouse, 
Observatory, May 1890, p. 179. 

' The whole neighbourhood of this cluster was perceived to be nebulous at 
Harvard College in 1870 (Annals, vol. xiii. p. 76). 


a thin, nebulous residuum,* which may, with a longer exposure 
than sixty minutes, be rendered unmistakable. Nebulae which 
thus linger on in clusters can scarcely be made manifest 
otherwise than with difficulty. For if, as we may be per- 
mitted to suj)pose, the stars replace their original more brilliant 
knots, what survives will usually be of the last degree of 

The spectra of all clusters are, in the main, continuous, 
but give signs of possessing individual peculiarities, a systematic 
scrutiny of which might profitably occupy some one of the 
great telescopes now in existence. Dr. Huggins was struck, 
in 1866, with the absence of red rays from the analysed light 
of the great cluster in Hercules, and perceived in it irregularities 
due either to bright or dusky bands.^ They were construed 
in the latter sense by Vogel in 1871,^ since when the inquiry 
has been unaccountably neglected. 

* Monthly Notices^ vol. xlviii. p. 30. 

« 1 hiL Trans, vol. clvi. p. S89. ' Astr. Nuch. No. 18C4. 




The fantastic variety of nebular forms was long a subject of 
wonder, scarcely tempered by a speculative effort. Inchoate 
worlds, disclosed with astonishing profusion by Herschers 
telescopes, seemed like mere * sports of nature ' in the sidereal 
spaces. Nebulae were to be found in the semblance of rings, 
fans, brushes, spindles ; they abounded in planetary, cometary, 
elliptical, branching varieties; nebulous shields, embossed 
with stars, or tasselled like the aegis of Athene, displayed 
themselves, as well as nebulous discs, rays, filaments, trianglea, 
parallelograms, twin and triple spheres. One nebula, thought 
to resemble the face of an owl, was named accordingly; 
another suggested a crab ; a third a swan ; a fourth (the great 
Orion formation) became known as the Fish Mouth nebula, 
from its supposed likeness to the gaping jaws of a marine 
monster. Fancy ranged at large through this wide realm, 
attempting to familiarise itself with the strange objects con- 
tained in it by finding for them terrestrial similitudes. 

Within the last few years, however — indeed, it may be 
said, since the completion of the Eosse reflector in 1845 — 
nebular inquiries have entered upon a new phase. A * glim- 
mering of reason ' has begun to hover over what long appeared 
a scene of hopeless bewilderment. With improved telescopic 
means — above all, with the aid of photography — structure has 
become increasingly manifest among aU classes of nebulae; 
structure, not of a finished kind, but indicating with great 
probability the advance of formative processes on an enormous 
scale, both as regards space and time. Masses that seemed 
all but amorphous when imperfectly seen, show to a keener 
scrutiny nodes and nuclei of condensation ; curving lines of 


light, telling of the presence of movement and force, furrow 
them; they are perceived to be rifted as if by a colossal 
thunderbolt, or riddled as if by a portentous cannonade. 
Simple milky effusions prove to be far less common than had 
been supposed, and excessive complexity of constitution is 
already a recognisable characteristic of most nebulfe. 

It is one which adds greatly to the interest of their study. 
For as the curious details of their organisation are laid bare 
by the intricate inequalities of their light, the prospect grows 
hopeful of gaining some insight into the nature of the systems 
formed by, or in preparation from them. Optical discoveries, 
while gradually acquiring physical significance, are helping to 
lay the foundation of a * nebular theory' emanating from 
augmented knowledge, and the discreetly adventurous thoughts 
which it may be supposed to countenance. 

Meanwhile, some mode of nebular classification has to be 
adopted for the guidance of our ideas ; and since their rapid 
modification through fresh detections allows no arrangement 
to be at present more than provisional, it will be best to depart 
as little as possible from that already in use. We may, then, for 
descriptive puri)oses, divide nebulaB into the following eight 
classes, which, nevertheless, frequently overlap so widely as to 
be barely distinguishable: 1. Nebulous stars. 2. Planetary 
nebulae. 3. Annular nebulae. 4. Cometary nebulae. 5. Spiral 
nebulae. 6. Double nebulae. 7. Elliptical nebulae. ■ 8. Irre- 
gular nebulae. 

In the course of one of his ' reviews of the heavens,' Sir 
William Herschel discovered a star in Taurus * perfectly in 
the centre ' of a * faintly luminous atmosphere ' about 3' in 
diameter.^ The consideration of this object (N. G. C, 1614) 
and of some others like it led him in 1791 to the memorable 
conclusion that there exists in sjDace * a shining fluid of a 
nature totally unknown to us.' Nothing, indeed, could be 
clearer than that * the nebulosity about the star was not of a 
starry nature,' and there is just as little doubt that it is no 
atmosphere in the ordinary sense of the word. This is de- 
monstrated by its extent alone. For the 'glow' round 

• Phil. Tram, vol. Ixxxi. pp. 71, 82. 


Herschers pattern nebulous star fills a sphere at least thirty 
times as wide, or 27,000 times as capacious, as that enclosed 
by the orbit of Neptune ; and the Eosso telescope disclosed 
irregularities of illumination within it,* suggestive of unfold- 
ing activities, directed, perhaps, towards the production of a 
planetary scheme far vaster and more elaborate than our 
own. Thirteen nebulous stars were enumerated by Herschel, 
and many have since been added to them. A fine specimen 
in Eridanus was picked up by Swift in 1859 ; * a small 
star in Canes Venatici came out strongly * burred ' on one of 
Mr. Eoberts's plates in 1889 ; ^ and among Sir John Herschel's 
southern discoveries was a close, sharply-defined double star 
surrounded by a bright luminous * atmosphere ' 2' in extent * 
(N. G. C, 5367). The glow formerly apparent about 55 
Andromedee seems to have vanished. The star, which is of 
5^ magnitude, was marked 'nebulosa' by Flamsteed and 
Piazzi ; * Sir John Herschel regarded it as an especially fine 
example of that peculiarity, and his description was corro- 
borated by Dr. Huggins in 1864 ; ® but Lord Eosse examined 
the object eleven times without finding a trace of nebulosity ; ^ 
and d' Arrest, Schjellerup, as well as Burnham with the 
eighteen-inch Chicago achromatic, were equally unsuccessful.® 
Another suspicious case of the same kind is met with in the 
still brighter star 8 Canum Venaticorum, perceived four times 
by Sir John Herschel, but by no one else, before or since, to 
be encompassed with a * considerable atmosphere.' ^ Photo- 
graphy might usefully be employed to clear up the doubt as 
regards both stars. Illusory glows cannot impose upon the 
sensitive plate. 

A * nebulous star' proper forms the centre of an ill- 
defined aureola ; but nebulous adjuncts to stars exist in every 
variety of branches and chevelures, wisps and whorls. In- 

• Trans. R. Dub. Soc. vol. ii. p. 40. ' Sidereal Messenger, vol. iv. p. 89. 

• Monthly Notices, vol. xlix. p. ?63. * Ca2W Observations, pp. 23, 107. 

• According to Schjellerup (Astr. Nach. No. 1613), Piazzi merely copied 
from Flamsteed. 

• Phil Trans, vol. cliv. p. 442. ' Travs. R. Dub. Soc. vol. ii. p. 60. 

• Memoirs R. Astr. Soc. vol. xlvii. p. 220. 

• Phil. Trans, vol. cxxiii. p. 427. 


deed, the sequence is so continuous between bright stars with 
filmy appendages, and pronounced nebulae involving minute 
stars, that it is often diflBcult to say whether the stellar or the 
nebular character predominates. Thus, 'planetary' nebulaB 
have often stellar nuclei ; they can be discriminated, however, 
from nebulous stars, first, by their mainly gaseous spectra, 
next,, by the comparatively definite termination of their discs. 

It is no wonder, then, if among Herschel's nebulous stars 
one was found not strictly entitled to bear that name. This 
nondescript object (N. G. C, 2392) is situated in Gemini, and 
that it struck him as sometliing unusual may be inferred from 
his designating it * one of the most remarkable phenomena I 
have ever seen.' * With the Parsonstown reflector it presented 
a * most astonishing ' appearance. Herschel's ' equally 
diflFused nebulosity ' was replaced by several bright and dark 
rings, varying in breadth, and perhaps spiral in their arrange- 
ment.^ The genuine nature of the combination might even 
yet be in doubt but for the dictum of the spectroscope. 
D' Arrest' found itsHght to be concentrated in the unidentified 
green ray of wave-length 5005— the central star or nucleus 
asserting its superior condensation by the display of a faint 
continuous radiance. It is then essentially a nebula, and 
generally passes for one of the annular, or perforated kind. 
A truly nebulous star of a reddish colour makes with it (not, 
we may surmise, fortuitously) an open pair at 105".'* The 
diameter of the system of rings is about 53'\ 

A somewhat analogous object, 30'' across, occurs in 
Cygnus (N. G. C, 6826). ' It is of a middle species,' Herschel 
wrote in 1802, * between the planetary nebulse and the nebulous 
stars, and is a beautiful phenomenon.' ^ To its greenish-blue 
disc and lucid centre corresponds a mixed spectrum of bright 
lines and uniformly dispersed light. 

Planetary nebulsB were first distinctively adverted to by 

» Phil Trans, voh Ixxxi. p. 81. 

« Trans, R. Dub. Soc. vol. ii. p. 59 ; see also H. C. Key, Mo7ithly Notices, 
vol. xxviii. p. 154. 

* Astr. Nach. No. 1896 ; Abhandlungeiif Leipzig, Bd. iii. p. 321. 

* Lassell, Memoirs R. Astr. Soc. vol. xxxvi. pp. 42, 61. 
» Phil. Trans, vol. xcii. p. 522. 


Sir William Herschel. Their classification caused him a good 
deal of perplexity. *We can hardly suppose them/ he re- 
marked at starting, * to be nebulae ; their light is so uniform 
as well as vivid, the diameters so small and well-defined, as 
to make it almost improbable they should belong to that 
species of bodies.' After, however, he had weighed and 
found wanting the hypotheses of their being actual planets 
belonging to distant suns, or distended stars, or comets near 
aphelion, he at last decided — rightly, as usual — in favour of 
their nebulous nature.^ 

Fifty of them were known when Pickering began 'sweeping 
in 1881, for * stars with remarkable spectra ; ' ^ and within 
a few years, upwards of twenty more were identified, through 
the quality of their light alone, by him and Dr. Copeland. 
These are, however, for the most part totally devoid of the 
visible disc which was the original badge of their class ; they 
are either very small, or very remote planetaries ; and are 
often distinguished as * stellar nebulae.' 

A true * planetary ' aspect has not, indeed, in any case 
survived the scrutiny of modern observers. What had seemed 
equably illuminated discs are broken up by the powerful tele- 
scopes now in use, into brighter and darker portions, dis- 
tributed in evident relation to some unknown conflict of forces. 
Some of these discs include strongly-marked nuclei ; others a 
sprinkling of minute stars ; condensation towards a spherical 
surface gives to many the aspect of a ring-shaped enclosure ; 
few (if any) are clean at the edges. 

The effects of progress in seeing may be exemplified by 
the history of a noted planetary nebula (M 97) discovered by 
Mechain, near y8 Ursee Majoris, the second ' pointer,' Febru- 
ary 16, 1781 . To both Mechain and Messier it appeared a barely 
discernible spot of faint light ; ^ but Sir John Herschel de- 
scribed it as * a most extraordinary object — a large, uniform 
nebulous disc, quite round, very bright, not sharply defined, 
but yet very suddenly fading away to darkness.' * At Parsons- 

> Phil. Trans, vol. Ixxv. p. 265. 

» Observatory, vols. iv. p. 81, v. p. 294 ; Monthly Notices, vol. xlv. p. 91. 

» Conn, des Temps^ 1784, p. 265. * PhU. Trans, vol. cxxiii. p. 402. 


town, in 1848, two stars* were perceived in the interior, each 
surrounded by a dark space encroached upon by nebulous 
whorls ; and the object received the name of the * Owl nebula,' 
from the appearance of two great oculi thus presented. 
Its dimensions were found to be 163' by 147"; but the out- 
lines of the disc seemed 'ragged,' its torn edges serving, in 
Professor Alexander's opinion, as * marks of disruption and 
dispersion outward.' * 

The * Owl,' like every other planetary nebula yet examined, 
gives a spectrum of bright lines. In the main, then, it is a 
globular mass of hydrogen and other gases of such incon- 
ceivable size that literally thousands of solar systems could be 
accommodated within its bulk. If we suppose it to be sixty- 
five light years distant from the earth — and the probability that 
it is much more remote comes nigh to certainty — its diameter 
must exceed that of the orbit of Neptune upwards of one 
hundred times ! Here, indeed, there is room and 

/^^\ verge enough for the unfolding, in a dim futurity, 

^^^^K of vast creative purposes. 
jHV.i This object might be called a })erf orated plane- 

^ y^ ^ tary nebula; others seem multiplex. Two or 
three superposed discs are traceable in them, 
Plan^^ Sketch recalling the complicated series of envelopes, of 

of a Nebula which the heads of many comets are con- 

^ °^^^* structed. A small oval planetary, for example, 
in Taurus Poniatowski (N. G. C, 6572), was resolved by Vogel 
with the great Vienna refractor, in 1883,^ into three strata of 
nebulosity disposed as in fig. 35, representing, no doubt, suc- 
cessive spherical and ellipsoidal envelopes of diminishing 
luminous power. 

An object (N. G. C, 6210) with an ' intense blue centre 
fading off to some distance all round,' and hazy at the edges,* 
was perceived by Vogel as triple. A faint oval husk (so to 

' The fainter star disappeared in April 1850, perhaps through the tarnishing 
of the mirror, Phil. Traris. vols. cxl. p. 513, cli. p. 721 ; Trans. R. Dub. !Soc. 
vol. ii. p. 93. 

* Astr. Jour. vol. ii. p. 141. ' Potsdam Puhlkationcn, No. M, p. 34. 

* Trans. /?. Dxib. Sec. vol. ii. p. 150. 


speak) seemed to enclose a vivid l-ernel, and that again to in- 
clude a stellar nucleus. This nebula is situated in the con- 
stellation Hercules; and one of a similar character in Eridanus 
(N. G. 0., 1535) struck Mr. Lassell as made up of a faint 
circumferential disc surrounding a brighter one which included 
a speck of condensed (but it was thought) non-stellar light. 
The whole effect was * extraordinary and beautiful.' * 

Perhaps the most interesting of all the planetary nebulae is 
one lying quite close to the pole of the ecliptic, near the star 
a) Draconis (N. G. C, 6543), also known as H iv. 37, or the 
thirty-seventh of Herschel's fourth class. Its longer diameter 
(for it is slightly elUptical) measures about 80"; it is of a blue 
colour, and shows a white star of eleventh magnitude giving 
a perfectly continuous spectrum exactly in the middle of a disc 
from which a purely gaseous one is derived. Dr. Huggins's 
first experiment in the analysis of nebular light was in fact 
made upon the planetary in Draco,^ which has in various ways 
been used as a test object. Attempts to determine its parallax 
were vainly made by d' Arrest, Briinnow, and Bredichin.* For 
proper motion, too, it was tried by d'Arrest in 1872, with a 
similarly negative result. During the eighty-two years elapsed 
from a careful observation by Lalande in 1790, the nebula 
had remained to all appearance completely stationary. But 
this fixedness was really to some extent communicative as 
regards its minimum distance from the earth. D' Arrest 
showed that unless this exceed a light-journey of forty-seven 
years, the nebula must have become sensibly displaced in the 
course of eighty-two years by the simple perspective effect of 
the sun's advance at the rate of five miles a second.^ And 
since there is little doubt that this estimate of the solar velo- 
city ought to be trebled, the least admissible distance of the 

> Memoirs R. Astr. Soe. vol. zzztL p. 40. 

• Popular Hist of Astr. p. 437, 2nd ed. 

' Briinnow obtained for H iv. 37 a nominal parallax, but Brediohin, from 
nearly doable the nmnber of observations, derived a negative one implying the 
nebula to be more remote than the tenth magnitude star with which it was com- 
pared. A$tr, Nach, Ko. 2916 (Oudemans) ; Eng, Mec. vol. xliii. p. 504 (H. 

« Asir, Nach, No. 1885. 


nebula should also be trebled. Its light then can only reach 
our eyes after an interval of 140 years,' 1/ then ; for it may 
spend a great deal longer on the road. The real size of the 
globe it emanates from mast be vast in proportion to such 
extreme remoteness. We find, accordingly, that at least forty- 
four diameters of the orbit of Neptune would be needed to 
measure it from side to side. 

Becent observations, however, raise a question as to 
whether the figure of the Draco planetary is even approxi- 
mately globular. With the great Lick refractor, Professor 
Holden perceived the disc to be made up of two superposed 
rings, one lying (it was inferred) behind the other -ini space) 

and generated either.4>reviously 
or subsequently to'^it in time 
(see fig. 36). Their arrange- 
ment, in short, seemed to be that 
of a true helix, and the object 
was hence ranked as ^ the first 
member of a class da^mibated 
•helical nebul»J^ The great 
merit of this. (Prions discovery 

Fig. 86.— Planetary Nebula resolved . . , . ., x i_i- i. 

into HeUcal Form (Holden). consists, not m the establish- 

ment o/^a new and barren dis- 
tinction, but in the rationalising of one already made. Helical 
nebulae may then be regarded as spiral nebul» under a novel 

A blue, or greenish tinge is more or less characteristic of 
all gaseous nebulae ; it is especially conspicuous in a planetary 
(N. G. C, 3918) discovered by Sir John Herschel in the Centaur, 
and described as * very like Uranus, only half as large again,' 
its * colour a beautiful rich blue, between Prussian and ver- 
diter green. '^ The light of this object is about equal to that 
of a seventh-magnitude star. 

A considerable number of planetary nebulae, as we have 
already partly seen, manifest both annular and spiral ten- 
dencies. In some, a marginal brightening gives, with suflS- 

' Corresponding to a parallax of 0"-023. 

» Monthly Notices^ vol. xlviii. p. 388. » Cape Observations, p. 100. 


cient telescopic power, the effect of a ring ; in others, curying 
lines of light betray incipient spirality of conformation. 
Since they partake of the nature of all the three species, their 
classification is to a great extent arbitrary. Five of Herschel's 
planetaries assumed in fact at Parsonstown a ring-shape.' 
One of these (N.G.C., 2438), the nebula involved in the cluster 
M 46,^ was observed not alone to be pierced with a nearly central 
cavity, but to contain two, perhaps three stars, towards one of 
which the exterior nebulous ring wound spirally inward.* A 

* hole ' too disclosed itself in a planetary in Gloria Frederici 
(N. G. C, 7662), observed by Lassell as bi-annular, ' a ring 
within a ring.' To Father Secchi, it had, with high powers, 
the effect of a * magnificent horseshoe of scintillating points,*^ 
the gutter of which was also evident to Vogel.* Yet the 
object is of a purely gaseous constitution, emitting the 
ordinary nebular trio of bright lines, to which, by a peculiarity 
shared with only two other known nebulae, it adds the blue 
ray characterising the spectrum of stars like y Argus.^ The 

* bi-annular ' planetary is either hazy or ' fringed ' at the 
edges, of a bluish colour, and measures 32'' by 28".'^ 

A * sky-blue likeness of Saturn * replaced, in Mr. Lassell' s 
reflector, a round, faintly lucent object (N. G. C, 7009) dis- 
covered by Herschel in 1782 near the star v Aquarii. With 
higher powers, the disc became a ring 26" by 16", hazy within 
and without ; and the whole interior assumed, under Vogel's ex- 
amination, a curious screw-shaped structure.® Professor Holden 
remarked an unexpected point of hkeness between this nebula 
and the one last mentioned (N. G. C, 7662), in the possession 
by both of an interior oval ring, singularly warped and twisted 
out of the central plane ; the peculiarity associating them also 
with the * helical ' planetary in Draco. The anssB, or handle- 

> Pha. Trans, vol. cxl. p. 607. * See ante, p. 249. 

• PkU. Trans, vol. cxl. p. 613. 

• Astr. Nach. No. 1018 ; Les EtoiUs, t. ii. p. 14. 

• Potsdam PublicatuyMn^ No. 14, p. 37. 

• Hugging, Phil. Trans, vol. cUv. p. 440 ; Harvard Annals, vol. xiii. p. 72. 
See ante, p. 78. 

' Lassell, Memoirs R. Astr. Soc. vol. zxxvi p. 61. 

• Potsdam Publ No. 14, p. 37. 

s 2 


like appendages, producing in the object near v Aquarii its 
resemblance to Saturn with half-opened rings, are a unique 
feature among nebulte. First represented by Lord Rosse, they 
were resolved by Professor Holden * into distinct luminous 
masses, just traceably connected with the main body (see fig. 37), 
and not improbably satellite-stars in course of formation. The 
attendance of small stars upon nebulse both planetary and 
annular, is too close and too constant to be accidental, and 

long ago attracted the attention 
^--^ of Sir John Herschel.' One 

^^- . \rF^ ^^ such group (N.G.C., 6818) struck 

^,,^__^ him as exactly like a planet and 

^s-tSpf^' pair of moons; and stars in slow 

Fio. 37.— Annular Nebula in transit across a nebula of which 

q U8( o en). ^^^^ ^^^ ^^^ dependents may 

tften appear projected upon it. But not the slightest evidence 
of movement in these ancillary objects has yet been detected. 
The typical annular nebula (M 57) was first noticed by 
Darquier of Toulouse in 1779, between ^ and 7 Lyrw. It 
consists of an oval bright ring, 80" by 60", the interior of 
which is filled with a dim nebulous haze like ' gauze stretched 
over a hoop.' * Harding, already in 1797, perceived irregula- 
rities in the illumination of the ring ; ^ and vivid patches are 
especially conspicuous at either extremity of the minor axis. 
Minima of light, on the '•ontrary, terminate the major axis ; 
the nebula, as photographed at Hereny, September 1, 1886, 
taking somewhat the shape of a pair of parentheses set a 
little apart, thus C , but with spiral links between.* With the 
Kosse reflector, filaments of nebulosity were seen streaming 
outward from the edges;** and to Father Secchi the ring 
appeared glittering with stars ' Uke finely powdered silver.' ^ 
Tempel too remarked upon the great number of stars visible 
without and within the nebula ; and Professor Hall, at Wash- 

' Monthly Notices^ vol. xlviii. p. 391. * Phil. Trans, vol. cxxiii. p. 600. 
« Sir J. Herschers Outlines of Astr, p. 644, 9th ed. 

* Holden, Monthly Notices, vol. xxxvi. p. 64. 

* Von Gothard, Astr. Nach. No. 2749. • Phil. Trans, vol. cxxxiv. p. 822. 
' Webb, Cel. Objects, p. 347. 


ington in 1877, perceived it to be ' surrounded by a ring of 
faint stars/ of twelfth to fourteenth magnitudes.^ 

Views of this object afforded to Professor Holden by the 
Lick telescope in 1888* included a bewildering multitude of 
sharp stellar points, arranged, with evident reference to the 
nebulous oval in their midst, into elliptical chaplets clinging 
to its inner and outer edges, neither of which was bounded by 
a smooth curve. The perspective relations of these appearances 
form a very curious problem. To us they present themselves 
as a flat picture, but of course deceptively ; we find it, however, 
difficult to realise the disposition of parts in the solidity of 
space, to which they truly correspond. The ellipfcicity of the 
ring, for instance, can scarcely be due to foreshortening ; since 
if it were, the marking of the least diameter by maxima, of the 
greatest by minima, of luminosity, should be a piure coincidence, 
which it evidently is not. 

A central star, discerned in this nebula by von Hahn at 
Bemplin, towards the close of the last century, was missed 
by him in 1800,* and has often since evaded the scrutiny of 
better-provided observers. It appeared on the Hereny and 
Liverpool photographs,^ but left no trace on plates exposed at 
Paris. Satisfactory proof, however, of its suspected variabiUty 
is still wanting. 

A nebula in Cygnus (N. G. C, 6894) might be called a 
reduced copy of that in Lyra. It measures 47'' by 41", the 
interior vacuity, which is partially filled with faint Ught, 20". 
A conspicuous star is included within it.* An object of the 
same kind in Scorpio (N. G* C, 6887) was described by Sir 
John Herschel as * a beautiful delicate ring of a faint ghost- 
like appearance, about 40" in diameter.'^ Two stars, or 
nebulous nodes, are placed in it diametrically opposite to each 
other, and the whole aspect of the nebula suggests extreme 

» Aatr. Nach, No. 21S6. 

' Monthly Notices, vol. zlviii. p. 388. " Astr, Jahrbuch, 1802, p. 10. 

* Von Gothard, Astr. Nach. No. 2749; Spitaler, ibid. No. 2800. 

* Lord BoBse, Trans. B. Irish Acad. vol. ii. p. 156. 
' Cape Observations, p. 114. 

' Lassell, Memoirs R. Astr. Soc. vol. xzzvi. p. 47 » 


Ify on the other hand, vicinity can be inferred from great 
apparent extent, Mr. Barnard's nebulous ring in Monoceros 
must be comparatively near the earth. Its outer diameter, as 
mentioned in the last chapter, was estimated at 4(y, its inner 
at 20^; and the brightest among several 'knots' apparent 
in it had been detected by Mr. Swift as a separate object 
(N. G. C, 2287) so long ago as 1865. Variations both of its 
shape and brightness are, indeed, surmised by him to be in 
progress.^ Strangest of all, there are indications that the 
formation really exists in duplicate. What seemed to be a 
section of another vast ring was perceived by Mr. Barnard 
close to, and perhaps in nebulous connection with, the first.' 
Their further investigation, especially by photographic means, 
may lead to very curious results. 

Four ring-nebulffi, two in the northern, two in the southern 
hemisphere, were known to the Herschels ; and, as we have seen, 
many so-called ' planetaries ' show annular, as annular nebulas 
show spiral procUvities. Rings, in some instances, visibly curve 
inward towards a nucleus, giving rise to the variety which we have 
designated ' cometary' nebulae. Thus, a ninth-magnitude star 
with nebulosity attached (N. G. C, 1999) was found at Parsons- 
town to resemble ' a comet coiled into a ring,* ' and was photo- 
graphed, precisely under the same aspect, by Mr. Common in 
1888.^ The triple star, i Orionis, less than a degree distant, is 
enveloped in a nebula of analogous form, which belongs also, 
in a measure, to the appendage of the star Maia in the Pleiades. 
Sir John Herschel's ' falcated ' nebulas are of the same kind. 
One such in Argo, 10' in extent (N. G. C, 3199), appeared to him 
of a 'semi-lunar shape,' diffuse outside, but with a sharp inner 
edge.* Another (N. G. C, 846) occurs far to the south in 
Hydrus. * A complete telescopic comet,' a perfect miniature 
of Halley's,® was encountered in the constellation Eridanus, 
(N. G. C, 1825), and star-like condensations, with brush or fan- 
like appurtenances, are not unfrequent on his lists. 

The discovery of spiral nebulae was beyond question the 

> Sidereal Messenger, Jan. 1890. « Astr. Nach. No. 2918. 

* Trans. R. Dub. Soc. vol. ii. p. 60. * Observatory, vol. xii. p. 84. 

* Cape Observations, pp. 20, 94. • Ibid. p. 61. 


most important result of the construction of the great Parsons- 
town reflector. Its significance is continually enhanced as the 
wide prevalence of convoluted forms among this whole class 
of sidereal objects is rendered more fully apparent by the 
increasing advance of exploration. 

The typical spiral nebula in Canes Venatici (M 51) pre- 
sents, with a great telescope, a truly amazing appearance. The 
two nuclei separately catalogued by Sir John Herschel are 
then seen to be connected by an exterior faint sweep from an 

Fio. 3S.— Spiral Nebula in Canes Venatici. 
{From a photograph taken at Her&ny.) 

inner system of wreathing nebulous bands. These, too, show 
nodosities and angularities * (well shown in fig. 38, from a 
photograph by M. von Gothard) in obvious mutual relations, 
as if the knots, instead of simply forming upon the spires, 
had determined, or at least deflected their course. A still 
clearer knowledge of their arrangement has been gained 
through a remarkable photograph taken by Mr. Koberts with 
four hours' exposure, April 28, 1889.* The nebula displays 
itself in it, no longer coiled like a watch-spring, but as com- 

> Vogel, Puhlvcationeriy No. 14, p. 32. * Monthly Notices, vol. xlix. p. 38. 



posed of a pair of curving arms issuing from opposite extre- 
mities of an oval central body. One of these loses itself in a 
vague efifusion, as a comet's tail dies out into darkness ; the 
other attains the secondary nucleus, and there terminates. 
The spiral character of this great vortex is perhaps rendered 
exceptionally conspicuous by its being more favourably placed 
than most others for our inspection. We seem to get nearly a 
bird's-eye view of it, and are thus enabled to take in the design 
of its construction at a glance. Its spectrum is continuous. 

Fig. 89.— Spiral Nebula in Virgo. (From a photograph taJcen at Heriny.) 

An object in Virgo 3' across (M 99) is a left-handed spiral; 
its branches turn the opposite way from those of M 51. Their 
tendency to form knots and angles is strikingly shown in fig. 89, 
copied from a photograph obtained in two hours by M. von 
Gothard, April 12, 1888.* A diffuse nebulous mass in Tri- 
angulum (M 83), just discernible with the naked eye, was shown 
by the Kosse reflector to be a 'large spiral, full of knots.'* 

Vogel, Astr. Nach. No. 2854. 

« Phil Trans, vol. cli. p. 711. 


Its light is of a continuous nature,* although it may not prove, 
on further examination, to be genuinely steUdr. 

Curved furrows of light, such as it is agreed to call 'spiral,* 
have been traced in many planetary and annular nebulsB; 
with still greater optical power than is now at the disposal of 
astronomers, they might possibly, be brought to view in all. 
The kind of structure which they indicate seems indeed to 
characterise, in some degree, every form of cosmical agglomera- 
tion, and depends no doubt upon laws of their ordained develop- 
ment foreign to our terrestrial experience. 

The tendency of winding nebulous bands to become knotted 
often proceeds so far that' the knots all but completely absorb 
the bands. Multiple groups of nebulsB then appear ranged 
along curved lines, the intermediate faint luminosity becoming 
perceptible only with large telescopic apertures. Such is the 
curious double nebula in Perseus (M 76), noticed by the pre- 
sent Lord Bosse to constitute, with subordinate nodules and 
streamers, a system modelled on a ' reaping-hook * pattern.^ 
A gaseous spectrum is derived from it. Similar combinations 
are met with in the southern constellation Mensa (N. G. C, 2046) 
where five nebulae are disposed along an oval line, and in a 
* falcated ' nebula with three knots situated in Cepheus (N. G. C, 

All the diversities of double stars, it was pointed out by 
Sir John • Herschel,' * have their counterparts in nebulee ; 
besides which, the varieties of form and gradation of light in 
the latter afford room for combinations peculiar to this class 
of objects.' Its members are surprisingly numerous, one in 
sixteen of the 5079 nebulce catalogued by Herschel in 186 i 
being in unmistakable connection with other adjacent objects. 
Double and triple nebulsB are usually spherical, centrally 
condensed, and with traces of mutual coherence. Their 
separation is thus still visibly incomplete ; orlntal revolutions 
can scarcely be assumed as probable ; nor is there any sign of 
their being in progress. 

* Harvard Annaht toI. xiv. p. 288. The nebula is No. 242 of the Harvard 
Photometric Catalogue. 

« Trans, R. Dub. Soc. vol. ii. p. 21. • Outlines of Astronomy, p. 647. 


A nebula in Ursa Major (N. G. C, 3690) divided by Swift in 
1885,^ makes probably the closest pair known ; and a carious 
reproduction, with greatly widened spatial intervals, of star 
systems like that of 7 AndromedsB occurs in a triple nebula 
in Virgo consisting of a bright round nebula, attended, at a 
distance of 5', by an extremely faint one which is itself double * 
(N. G. C, 6813-14), Another compound object of a striking 
character was noticed by Sir John HerscheP in Canes 
Venatici (N. G,C., 4631), where an enormously long ray of 
nebulosity has a round, dimly luminous companion, a tenth- 
magnitude star placed between serving perhaps as a centre 
of attraction for both. 

A very close double nebula in Gemini (N. G. C, 2371-12) has 
also an intervening star symmetrically located in the line 
joining their centrea.^ Cirrus-like streaks of nebulosity 
partially encircle the two objects. Duplicity is, in other 
cases, still less clearly defined. Thus, a pair of nebulas near 
7 Leonis (N. G. C, 3226-27) are together enclosed in a faint 
luminous envelope, the effect recalling that of the celebrated 
* Dumb-bell ' nebula in Vulpecula (M. 27),* which is only 
perceived to be essentially single when the ' neck ' uniting 
two conspicuous hazy masses is brought into view with a 
powerful telescope. Sir John Herschel first observed the 
elliptical outline of the entire to be rounded out by faint 
luminosity, and thus saw it in its true aspect as a large, 
diversified oval disc, measuring about 5' by 8'. It might 
indeed be called a magnified planetary nebula not devoid of 
annular inclinations. The possibility that, by the progress of 
the central contraction and marginal spreading indicated by 
its present hourglass shape, the chief part of its mass may, 
in the course of ages, become diffused into a ring, is strongly 
suggested by the analogy of the bright spots at either end of 
the minor axis of the ring-nebula in Lyra. A planetary in 
the southern hemisphere (N. G. C, 1365) appears, in fact, to 

* Sid. Mess. vol. iv. p. 39. 

« D' Arrest, Astr. Nach. No. 1369. • Phil. Trans, vol. cxxiii. p. 431. 

* Lassell, Memoirs R. Astr. Soc. vol. xxiii. p. 62 ; Lord Bosse, PhU. Trans. 
vol. cxl. p. 612. 

* D'Arrest, Abhandhmgen, Leipzig, 1867, p. 325. 


have already reached a more advanced stage on the same 
road, and several ' miniatures * of the * Dumb-bell * are 
included among that class of objects. One especially in 
Cygnus (N. G. C, 6905), depicted by Vogel with the Vienna 
27-inch, easily gives an impression of actual duplicity,^ and 
showed at Farsonstown as a ' beautiful little spiral/ It has 
a central star, and four ' satellites/ Its spectrum, like that 
of the Dumb-bell nebula, is (so far as is yet known) absolutely 
monochromatic. The leading nebular ray at 6005 concen- 
trates the whole of its light. 

In a photograph of the Dumb-bell nebula, taken by Mr. 
Roberts with an exposure of three hours, October 3, 1888, the 
shaping-in towards the middle, so marked to the eye, is 
almost obliterated through the prolonged accumulation of 
chemical effects ; but it intimates pretty clearly the approxi- 
mate completeness of the oval bright border of the disc, as 
well as its superposition upon a fainter, more eUiptical one, 
visible as a kind of effusion at the extremities of its longest 
diameter. Yogel's drawing too ^ suggests, though after a 
different fashion, the presence of two ellipses, one partially 
concealed behind the other ; and there hence seems reason to 
think that this singular formation partakes, in more ways than 
one, of the compound character evident in many planetary 

> Potsdam Publicationen, No. 14, p. 36. ' Ibid, p. 35. 




The elliptical and irregular classes of nebulae are illustrated 
by such splendid examples that we have thought it well to 
devote a chapter to their separate consideration. One member 
especially of each towers above the rest, like Ajax among the 
Argive host, its rival alone excepted, and the two are so 
diflferent that it is not easy to award the palm of superiority 
to either. Needless to say that we allude to the objects in 
Andromeda and Orion, the types respectively of the elliptical 
and irregular plans of nebular construction. 

The former (M 31) is the only real nebula which can 
readily be detected with the unaided eye, and it is the only 
one, accordingly, which was discovered in pre-telescopic times. 
Al Sufi was familiar with the ' Uttle cloud ' near the most 
northern of the three stars in the girdle of Andromeda ; * and 
its place was marked on a star-map brought from Holland 
to Paris by De Thou, and believed to date from the tenth 
century.^ Simon Marius, who was the first to turn a telescope 
upon it, December 15, 1612, called it 'stellam quandam 
admirandse figurse,' and compared its dull and palUd rays to 
those of a candle shining by night through a semi-transparent 
piece of horn. Yet this strange phenomenon was only rescued 
from neglect by BouUiaud, whose attention was directed to it by 
the passage of the comet of 1664 across that part of the sky. 
So surprising did the disregard of it by Hipparchus, Tycho, 
and Bayer then appear to him that he concluded it to vary in 
light, an hypothesis which, however, derives no support from 
recent observations. 

> Schjellerup, Description des fjtoUes, p. 120. 
« Le Gentil, Mimoires de VAcad. 1759, p. 459. 


With powerful light-concentration this ' most magnificent 
object ' (to borrow Sir John Herschel's phrase) assumes vast 
proportions. They were extended by G. P. Bond, using the 
fifteen-inch refractor of Harvard College, to cover an area of 
4° X 2^°, and he probably did not reach their absolute limits. 
Two adjacent nebulae, one (M 82) descried by Le Gentil in 
1749, the other (N. G. C, 205) by Caroline Herschel in 1783, 
undoubtedly fall within their compass.' ' The light of this 
nebula is ' of the most perfectly milky, absolutely irresolvable 
kind.' ' It does not collect into ' floccules,' and produces 
none of the scintillating effect giving to many gaseous 
nebulsB a delusive appearance of resolvability. From the 
circumference towards the centre, however, it gradually 
brightens, then abruptly condenses to a small nucleus of 
indistinct outline under high magnifying powers, and possibly 
(Uke the nuclei of many comets) granulated^ but assuredly 
not stellar. 

This progressive brightening inward shows, nevertheless, 
interruptions. On September 14, 1847, Bond discovered two 
long dark rifts running nearly parallel to one another, and to 
the axis of the nebula.' Their detection was a consequence 
of the widened area of the luminosity perceived by him, the 
inner rift having been taken, until then, for its boundary in 
that direction. The outlines of Bond's drawing are given in 
the accompanying diagram by Mr. Wesley (fig. 40), in which 
the ' rifts ' are marked A and B. C represents Le Gentil's, 
D Miss Herschel's attendant nebula, E an exceptionally 
lucent region crowded (it has since been found) with hosts of 
minute stars.^ 

These enigmatical appearances at last assumed an intelli- 
gible form in a photograph taken by Mr. Roberts, October 1, 
1888.* The view given by this magnificent picture of the 
Andromeda nebula as a symmetrical, though still inchoate 
structure, ploughed up by tremendous, yet not undisciplined 

> Bond, Memoirs Amer, Acad, vol. iii. p. 88. 

* J. HerBchel, Memoirs R, Astr. Soc, toI. ii. p. 496. 
■ Memoirs Amer. Acad. vol. iii. p. 80. 

* Banyard, Knowledge^ vol. xii. p. 76. 

. * Monthly NoticeSt toI. zlix. pp. 65, 120. ■ 



forces, working harmonioasly towards the fulfilment of some 
majestic design of the Master Builder of the universe, is of a 
nature to modify profoundly our notions as to how such 
designs obtain their definitive embodiment. Plate III. re- 
produces an impression secured by the same artist with four 
hours' exposure, December 29, 1888, Although not covering 
so large an area as the visual representation in fig. 40, it is 

Fio. 40. — Diagram of the Great Kebala in Andromeda, from Bond's drawing 

in 1S47. 

of a totally different order of merit as regards clearness and 
consistency of details, the significance of which can be at 
once perceived by reference (in fig. 41) to Mr. Wesley's careful 
tracing of the lines of conformation brought out on the 
sensitive plate. 

Bond's ' canals' are now seen prolonged and curved into 
two vast rings (AA and BB), which prove, on attentive con- 
sideration, to be dusky intervals separating the successive 
spires of a single great stream of nebulous matter, winding 
outward from near the primary to reach the secondary nucleus 

Plate 3. 


Photographtd by Isaac Roberts, F.R.A.S., December ^gth, 1888. Exposure four hours. 


(M 32). The similarity of the relations between the two nuclei 
here and in the ' whirlpool ' nebula in Canes Yenatici scarcely 
needs to be emphasised by a remark. Thousands of stars are 
scattered over and around the Andromeda nebula ; the situa- 
tion of which in a prolongation of the Milky Way perhaps 


.. ■ T^. 



) ': 


* • ^ 




1 -. , 

• -;•' A ' 


/ • 


\ •'• *» , "• • 

*•/• ' • 


•w- • 


* ' * • V 



■. ^ -^i.. 

vs \ 

V .\. 

:-'^A \ 

• A'. V\ ^ *^ 

\' "s--. 

. *;\ V\ 


•\ . ". '•• v> 



••• •. ••>. ^\ 




%. 1 


Fig. 41. —Diagram from Boberts's photograph of the Andromeda Nebula. 

sufficiently explains their profusion. Many of them may be 
entirely disconnected from it ; but not all. Shoals of sharp 
stellar points assume, in the photographs, the annular or spiral 
arrangement of the nebula, and * lie along the edges of the 
dark rifts following all their sinuosities.' ' 

* Banyard, Knowledge^ Feb. 1 and Aug. 1, 1889. 


A palpable discrepancy between the drawings and photo- 
graphs of this stapendoas object was adverted to by Mr. 
BobertsJ The small oval nebala marked with the letter D in 
fig. 40 lies, it is easy to see, mach more obliquely towards the 
principal centre of the formation than the corresponding ob- 
ject in Plate III. And since a delineation of it by Trouvelot in 
1876 varies from the photograph in the same way as Bond's, 
the change, if real, must have taken place very rapidly, and 
may be expected to continue. 

Nothing is at present certainly known as to the constitution 
of the Andromeda nebula. Mr. Lockyer infers it to be meteoric^^ 
and compares its state to that of ' a comet within a month of 
perihelion.* * That no genuinely continuous spectrum is 
derived from it * was perceived by Dr. Huggins in 1864 ; but 
the nature and origin of the bands and lines, possibly both 
bright and dark, composing its prismatic radiance have yet 
to be ascertained. Only a scanty supply of chemical rays is 
included within the hmited range (from about D to F) of the 
latter; so that the difficulty of getting an approximately 
complete picture of the nebula by photographic means seemed 
wellnigh insuperable until it was overcome. 

The real shape of the formation must be that of a disc, oval 
or round, but with a globular mass both at the origin and ex- 
tremity of the nebulous spires, the convolutions of which we 
perhaps see greatly foreshortened. Their extent in space we 
can only estimate by making a precarious assumption as to 
their remoteness. Taking the distance of the nebula, for 
instance, to be of sixty-five light years, that already attributed, 
for illustrative purposes, to the cluster in Hercules, we find its 
radius to measure 162,000 times the radius of the earth's orbit ; 
so that the frontier of this glimmering realm, as determined by 
Bond, is much more than half as remote from its centre as the 
nearest fixed star (a Centauri) is from ourselves ! In travelling 
from end to end of it, light spends nearly six years ; and if it 
lie obliquely towards us, then our view of the further margin 
may be of an earlier date than our view of the hither margin, 
by a couple of years or more. Extensive changes of luminosity 

* Observatory, vol. xii. p. 61. » Ibid. p. 98. * See ante, p. 81. 


tvithin the nebula might then manifest themselves to us suc- 
cessively, although they had really occurred simultaneously ; 
while, on the other hand, the coincidence in time to our senses, 
of widespread variations, would argue a position of the nebula 
nearly square to the line of sight. That it is still (so to speak) 
in the plastic stage, there can be little doubt ; and the outbreak 
of the ' new star * of 1885 has shown that the action of the 
powers engaged in moulding it to its appointed shape may 
occasionally be attended by a catastrophic hberation of 

Fig. 42.— Photograph of the Milky Way in Sagittarius. 

But what is *^ its appointed shape'? What is this mar- 
vellous system destined to become, as it emerges from the 
nebulous condition ? In framing a reply to such questions we 
have only analogy to guide us ; yet the presumption is very 
strong of a general similarity both of modes of action and of 
their results in every part of the sidereal world. Look, for 
example, at fig. 42. It represents, from an admirable photo- 
gi^aph by Mr. Barnard, a great collection of Milky Way stars 
in Sagittarius* The area covered by it somewhat exceeds that 
of the Greater Magellanic Cloud — forty-two square degrees — 


and the stars thronging it range to below the thirteenth mag- 
nitade. The fundamental method of their arrangement can 
at once be apprehended. They are evidently condensed into 
an oval naclear mass fenced round (as was pointed oat by 
Professor Holden) by a partially vacant ring. When the glass 
positive from which the figure has been printed is held so far 
from the eye that it begins to assume a nebulous appear- 
ance, not one alone, but several half-obUterated rings come 
into view, constituting a series of spirally-connected clearings, 
the result, apparently, of some prevalent law by which 
the distribution of the clustered stars is governed. Thus 
seen, the photographed group presents so striking a resem- 
blance to the self-depicted Andromeda nebula, that it is 
difficult to reject the evidence it affords of close relationship, 
or to deny the possibility that in the cluster we see the 
exemplar of what the nebula, at some future epoch, will have 

It may be added that the apparently connected stellar 
groups M 24 and M 17 may be, in some sort, the equivalents 
of the attendants upon the Andromeda nebula; while the 
cascades of stars falling to the left of the main cluster are 
nebulously represented in the object portrayed (from a photo- 
graph by Mr. Roberts) in Plate IV. 

This is an elliptical nebula in Ursa Major (M 81) which 
reproduces, on a small scale, most of the features of the Andro- 
meda nebula. Lord Bosse estimated its greatest extent at 
about 16^ but this limit is considerably transcended in the 
photograph,^ which brings out, in addition, the essentially 
spiral structure of the object. The presence of a two-fold 
system of whorls, like those forming the nebula in Canes 
Yenatici, is, in fact, strongly suggested by the impression on 
the sensitive plate, of two sweeping trains of nebulous matter, 
issuing tangentially from opposite extremities of the forma- 
tion, as if in continuance of two distinct original effusions 
from the nucleus. 

A companion nebula (M 82), half a degree distant from M 81, 
was described by Lord Bosse as ' a most extraordinary object, at 

* Roberts, Monthly Notices, vol. xlix. p. 362. 

Plate 4. 


Photographed by Isaac Roberts^ F.R.A.S., March 31st, 1889. Exposure 3i hours. 


least l(y in length, and crossed by several dark bands.' * It is 
bi-nuclear, of a double lance-head shape, and is doubtless an 
elliptical spiral viewed edgewise. Although it appears with its 
primary on Mr. Boberts's plate, its relegation to near the edge 
of the field has placed it at a disadvantage as regards defini- 
tion, and so the instructive particulars about it which might 
otherwise have been gathered, remain unrecorded. A nebulous 
star completing the group ' is perhaps physically connected 
with it. The continuous spectrum derived from M 81, 82, as 
from other elliptical nebulae, proves to be no less deficient in 
red light than the spectrum of the Andromeda nebula.^ We 
may then safely attribute an almost identical constitution to 
all the three objects. 

Sir John Herschel noticed an approach to an annular con- 
formation in many of this class of nebulse, and made the curious 
remark, that ' as the condensation increases towards the middle, 
the ellipticity of the strata diminishes.' ^ This looks as if their 
ovalness were not a purely visual effect, and other circum- 
stances tend to confirm the supposition. Thus, in cases where 
either the foci of a nebulous ellipse (N. G. C, 6595), or the ex- 
tremities of its major axis (N. G. C, 6648), are occupied by a 
pair of stars, it may be considered certain that the appearance 
of elongation corresponds to the reality. 

The longitudinal clefts often visible in ray-shaped nebulaa 
imply an emphatic assertion to- the same effect. For other- 
wise, why should they run lengthwise, rather than in any other 
direction ? If the bodies they characterise were, in fact, cir- 
cular discs, none of their physical features could have any 
relation to the appearance they happen to assume by projection. 
Such a relation is, however, extremely conspicuous in an 
elliptical nebula in the Centaur (N. G. C, 5128), divided along 
its entire length by *a perfectly definite, straight cut, 40" 
broa^.' A delicate nebulous streak runs between, and parallel 
to the halves, which are sharply bounded on the sided facing 
each other, but hazy on those averted. * The internal edges 
of this very problematic object,' Sir John Herschel remarked, 

> Trans. R. Dub. Soc, toI. ii. p. 79. * See ante, p. 258. 

* Huggins, PhiL Trans, vol. clvi p. 388. * Cape Observations^ p 22. 

T 2 


* have a gleaming light like the moonlight touching the outline 
in a transparency.' ^ It is by no means a solitary example. A 
long narrow nebala in Leo (N. G. C, 3628) is ' split into two 
parallel rays.' ' A black chasm, into which the nucleus pro- 
trudes, separates a lucid from a faint streak in Coma Berenices ' 
(N. G. C, 4665) ; and a nebula in Andromeda (N. G. C, 891), 
with * a chink in the middle, and two stars,' supposed by Sir 
John Herschel to be ' a thin flat ring, of enormous dimensions, 
seen very obliquely,' would rather appear, from the Parsonstown 
observations, to belong to the numerous category of cloven rays. 
That the distinctive peculiarity of these depends upon some 
general constructive principle cannot readily be doubted; 
but we should vainly attempt to speculate upon its nature. 

No descriptive formula is wide enough to include all the 
capricious forms of * irregular ' nebulsB. In regarding these 
singular structures, we seem to see surges and spray-flakes of 
a nebulous ocean bewitched into sudden immobility; or a 
rack of tempest-driven clouds hanging in the sky, momentarily 
awaiting the transforming violence of a fresh onset. Some- 
times, continents of pale light are separated by narrow straits 
of comparative darkness; elsewhere, obscure spaces are 
hemmed in by luminous inlets and channels. The 'great 
looped nebula' (80 Doradus), one of the inmates of the 
greater Magellanic Cloud, resembles a strip of cellular tissue. 
It serves not to conceal, but to ornament in a wide, openwork 
pattern the sky behind it, and was described by Sir John 
Herschel as 'an assemblage of loops,' the 'complicated 
windings ' of which constitute it ' one of the most extra- 
ordinary objects which the heavens present.' ^ 

The ' net- work nebula ' (N. G. C, 6995) is of a somewhat 
similar nature. ' A most wonderful phenomenon ! ' Sir John 
Herschel exclaims. * A very large space' (about 25' by 12') 

* full of nebulse and stars mixed. A network or tracery of 
nebulsB follows the lines of a similar network of stars.' ^ The 

* dark hollows left by the interwoven nebula * were perceived 

* Cape Observations, pp. 20, 105. 

« Trans. R. Dub, Soc, vol. ii. p. 95. » Ibid, p. 118. 

* Cape Observations, p. 32. » PhU, Tram. vol. cxxiii. pp. 469, 508. 


by Mason and Smith at Yale College, in 1839, to be almost 
bare of stars ; VihUe a map of those involved sufficed to show 
with tolerable accuracy the windings of the nebula.^ These 
observers ascertained, too (what Sir William Herschel had 
suspected),* its union with a * bifurcate, milky ray ' (N. G. C, 
6992) in the neighbourhood into an immense complex for- 
mation, two or three degrees long, the intricacies of which 
might probably be studied with greater advantage by photo- 
graphic than by visual means. 

But all other irregular nebulae sink into insignificance 
compared with that shown by an opera-glass as a silvery 
patch round one of the stars in Orion's sword. This extra- 
ordinary object (M 42) has been under effective observation 
for 234 years, and during the last seventy has been mono* 
graphed, mapped, measured, figured, and photographed with 
a diligence worthy of its pre-eminence. Hence, future changes 
in it, should they take place, must be sUght indeed, to escape 

The multiple star Orionis ^ might be called the foundation 
stone of the entire structure. All the Unes of its architecture 
are laid down with reference to it, and the intimate physical 
association of the stars with the gaseous stuff surrounding 
them has been spectrographically demonstrated by Dr. and 
Mrs. Huggins.^ This gaseous stuff, although it pervades the 
trapezium, seems less luminous there than elsewhere. The 
space about the stars usually forms a sort of oasis of com- 
parative darkness in the midst of a wilderness of piled-up 
flakes of light. Usually, not always. D'Arrest, it is true, 
invariably saw the stellar group on an almost black ground ; 
but 0. Struve several times, and especially in 1861, found the 
trapezium as densely nebulous as the contiguous tracts ; ^ and 
the same appearance was noted by both Schroter and Lament.® 

The brightest part of the nebula, called, from its first 
delineator, the 'Huygenian region,* is shaped like a right- 

» Trans. Amer, Phil, Soc. vol. vii. p. 178. * PhU, Trans, vol. Ixxv. p. 31. 
> See anUy p. 216. « See ante, p. 79. 

* MHnoires de VAcad. de St. Pitershourgj t. v. No. iv. p. 1 16. 

• Holden, Washington Observations, 1878, App. I. p. 100. 


angled triangle. Over this space the light is collected into 

* flocculent masses/ which, with the best seeing, prove to be 
throaghout of a 'hairy' texture. The effect was compared 
by Sir John Herschel to the * breaking-up of a mackerel-sky, 
when the cloads of which it consists begin to assume a cirrous 
appearance,' ' and suggested to Mr. Lassell ^ ' large masses of 
cotton-wool packed one behind another, the edges pulled out 
so as to be very filmy.' The ' pea-green colour ' of the whole 
object struck him forcibly, but is apparent to most observers 
as little more than a greenish tinge. 

Emanations (or what seem such) from the Huygenian 
core stretch away in wide curves to form the outlying portions 
of the nebula. One great effluence runs out into a ' proboscis ' 
attached to the upper jaw of the nondescript creature limned, 
in almost unearthly radiance, on the sky. Another, repre- 
senting the lower jaw, bounds on the northern side the chasm 
of the distended fauces (the ' Sinus Magnus '), and is sub- 
sensibly connected — as if by a shoal leading to an island — 
with the * nebula minima ' (M 43), a rounded mass which 

* appears as if just drawing together into a star.' * Behind, 
and between the two, misty effusions spread far afield; 
resolved by G. P. Bond into an intricate fabric of convoluted 
and branching filaments, the brighter region from which they 
spring displaying a similar, but more compact mode of aggre- 
gation. * It is now impossible,' he wrote after a particularly 
fine view of the nebula, on February 26, 1861, • to see it in 
any other aspect than as a maze of radiating, spiral-like 
wreaths of nebulosity, or filamentous tentacles, the centre of 
the vortex being about the trapezium.' * And to Mr. Safford, 
using the then recently constructed 18^-inch Clark equatoreal, 
it appeared as ' an assemblage of curved wisps of luminous 
matter, which, branching outward from a common origin in 
the bright masses in the vicinity of the trapezium, sweep 
towards a southerly direction, on either side of an axis passing 
through the apex of the Regie Huygeniana.' * 

• Meffioira R. Astr. ScC' vol. ii. p. 491. ' Ibid. vol. xxiii. p. 54. 
■ J. Herschel, Memoirs R» Astr. Soc, vol. ii. p. 495. 

* Harvard Antiahf vol v. p. 158. * Ibid. p. 169. 

Plate 6. 


PhotogrtLphtd hy IstuK Roberts. F.R.A.S., Decemhtr i^th, 1888. Exfosure 81 wtinuies. 



The application of photography to this amazing object 
has not only supplied records of its actual condition in- 
definitely more authentic than any producible by the human 
handy but has served to combine the peculiarities of its con- 
formation into a strikingly suggestive whole. Plates Y. and YI. 
exhibit two impressions obtained by Mr. Roberts, the first, 
and more pictorially perfect, with an exposure of 81 minutes, 

.. — * '• 


Fio. 43.— IndeX'Diagram to Structures photographed in the Orion Nebula. 

the second in 8 J hours, February 4, 1889. The unprecedented 
comprehensiveness of the latter has indeed been secured with 
some sacrifice of characterisation. The central tracts appear 
' dark with excessive bright ' — their particular features lost in 
the general glare. But the main outline comes out with 
unlooked-for definiteness, and approximate completeness. The 
result can best be judged of by reference to fig. 43, exhibiting 


an * index' diagram* prepared by Messrs. Banyard and Wesley 
from several negatives (all by Mr. Roberts) of varied exposures- 
Two points are made perfectly clear by it. First, that the 
whole fabric of the nebula is concave towards an axis * passing 
through the trapezium in a north-easterly and south-westerly 
direction.** Next, that the effluences from the trapezium 
have a predominant tendency to assume ramified, or tree- 
like forms. Thus the seemingly eruptive jet marked with 
the letter b mimics the shape of a stone-pine, and bears a 
less equivocal resemblance to the * stemmed ' type of solar 

Fio. 44. — Group of Synclinal Structures from a photograph of the Corona 
of 1871. Drawn by Mr. W. H. Wesley. 

prominences, as well as to certain arboreal structures 
visible in some photographs of the solar corona. The latter 
analogy is rendered still more apparent by comparing the 
vast nebulous out-growths, e and h, with the group of coronal 
rays represented in fig. 44. The same kind of structures 
seem, in both cases, at once to spring upward and to curve 
inward, as if under the influence of a two-fold action — 
outward from a centre, and inward towards an axis. This 
organic similarity— first detected by Mr. Kanyard — between 
the Orion nebula and the luminous appendages of the sun, 

' Ranyard, Knowledge^ vol. xii. p. 147. 

Plate 6. 


Pkotographtd by Isaac Roberts, F.R.A.S., February ^tA, 1889, Exposure aos msmutes. 

■ I . 

. •;, I 


is rendered more profound by the association of helium with 
hydrogen in the chemical composition of both. 

The Umits of the great ' sword-handle ' nebula are con- 
tinually being pushed further back, and there is no reason to 
beUeve the process nearly terminated. On Mr. Boberts's 
plates/ first the ^ nebula minima ' was joined on to the main 
body, then, with lengthened exposures, the cloudy mass to the 
north (N. G. C, 1977) shown near the bottom of Plate VI. 
was reduced to its true position of an offset, by which the 
forms of the parent-body are pretty closely imitated.* But 
even the combined object is far from representing the nebulous 
contents of this part of the sky. Over an area of 150 square 
degrees, in which it is nearly central, twelve new nebulae were 
photographed at Harvard College in the spring of 1888, and 
indications were obtained that Sir William Herschel's surmise' 
of the union into one immense stratum of the * great' nebula 
with others lying north and south of it might be verified in 
the immediate future.^ A spiral form for the stratum is 
further strongly suggested by * long-exposure * impressions 
secured by Mr. W. H. Pickering in California* and Peru. 

It may be that we see in the Orion nebula a great unde- 
veloped cluster on the model of the Pleiades. Nearly a 
thousand stars were catalogued by Bond in the portion of it 
(8*86 square degrees in area) examined by him, and his list 
seems to be practically exhaustive. Many of them should, 
indeed, in Professor Holden's opinion,® be accounted rather as 
nebular condensations than as true stars ; but that they will 
eventually grow to be such, as they slowly absorb the nebulous 
material now enshrouding them, is a justifiable conjecture. 
The same process appears already far advanced in the Pleiades, 
with the result that the stellar now altogether predominates 
over the nebular element in the compound system. It can 
scarcely have been always so. The balance inclined perhaps, 
in the remote past, as decisively the other way as it now does 
in the Orion nebula. 

» Monthly Notices, vol. xlix. p. 296. » Ranyard, loc, cit. p. 148. 

■ Phil. Trans, vol. Ixxix. p. 249. * Harvard Annals , voL xviii. p. 117 

» Hid, Mess. vol. ix. p. 1. • Washington Observations 1878 App. i. p. 221. 


It is not easy, Sir John Herschel tells ns, ' for language to ' 
convey a full impression of the beaaty and sublimity of the 
spectacle offered by the Argo nebula, when viewed in a sweep, 
ushered in as it is by so glorious and innumerable a procession 
of stars, to which it forms a sort of cUmax.' ^ ' Situated in 
one of those rich and brilliant masses, a succession of which, 
curiously contrasted with dark adjacent spaces, constitute the 
Milky Way between Centaur and Argo,' its branches with their 
included vacuities cover more than a square degree, and are 
strewn by above twelve hundred stars. The peculiarity, giving 
the formation the name occasionally applied to it of the ' key- 
hole ' nebula, is a large lemniscate-shaped opening in the central 
and brightest part, the blackness of which is qualified only by 
the veiling of one corner by a strip of thin nebulous haze. 
Four stars are placed precisely at the edges, none within the 
vacuity ; and the famous variable tf Argus ^ lies close to its 
eastern border. Such was the brilliancy of the star in 1838 
as almost to obUterate the ' key-hole.' now, and previously to 
the outburst, the special, individualising feature of what would 
otherwise seem a chaotic sea of luminous billows. The signi- 
ficance of the peculiarity is emphasised by its duplication. A 
second oval aperture, completely dark but for the faint sparkle 
of four minute stars, occurs in the southern, sparser part of 
the nebula. A similar echo of a characteristic form is met 
with, as our readers may remember, in the great Hercules 

A very peculiar gaseous nebula in Sobieski's Shield (M 17) 
has revealed itself piecemeal, and may still have disclosures in 
reserve. Messier, first of all, noticed there in 1764 a spindle- 
shaped, starless ray, about 6^ in length. Sir William Herschel 
added an arch, springing from its western extremity ; and the 
combined object became known as the ' Horseshoe ' or ' Omega ' 
nebula, its form resembling that of the Greek capital letter XI, 
with the left-hand base-line turned up obliquely.^ Again, it 
suggested a ' Swan ' to observers whose instruments were in- 
adequate to show the complete arch, and Flammarlon compared 

> Cap2 Observations^ p. 38. ' See anUt p. 116. * See ante, p. 246. 

* J. Herschel, PhiL Trans, vol. cxxiii. p. 461. 


it to a smoke-drift, fantastically wreathed by the wind. Signs 
of incipient organisation are, nevertheless, traceable in it. At 
the ' elbow,' or angle between the base and the swan-neck, 
Sir John Herschel detected two bright knots, tending evidently 
to become insulated from the surrounding nebulosity ; and 
perceived further, in the clear air of the Cape, a second very 
faint ^ horse-shoe,' attached to the opposite end of Messier's 
streak from the first.^ This new part of the nebula seems to 
have been independently re-discovered by Swift, of the Warner 
Observatory (N. Y.), July 4, 1888 ; * and both he and Herschel 
saw or suspected additional patches and convolutions. The 
edifice has yet to be * crowned ' with the assistance of the 

The nebula, as at present known, presents the outline 
rather of a two-arched bridge with a wide pier, than of 
a horseshoe. It appears to include two centres rotating, 
presumably, in opposite directions ; since, of the inflected arms 
of light originating from them respectively, one follows the 
lines of a right-handed, the other of a left-handed spiral. It 
is possible, too, that their connection is not designed to be 
permanent. The ' pier ' betweeen the ' arches * may eventually 
collapse. The conjecture at least is not unreasonable that, under 
the strain of counter-soUcitations, the continuity of Messier's 
streak will at last give way, and the matter contained in it be 
gathered round the pair of vortices constituting the nuclei of 
a double spiral nebula. 

The ' Trifid ' nebula in Sagittarius (M 20) has a dark 
central space. The three great lobes of which it consists 
are divided by * a sort of three-forked rift or vacant area, 
abruptly and uncouthly crooked, and quite void of nebulous 
Ught.'* These tracks of darkness are thormighfares. They 
run right through from the brilliant centre to the dim circum- 
ference of the nebula. A conspicuous multiple star,^ placed 
just at the sharp inner edge of one of the luminous detach- 
ments, probably exercises a predominant influence over them 
all ; and corresponding ascendency may be claimed for a single 

* Cape Observations^ p. 10. * Sidereal Messenger^ vol. iv. p. 38. 
* Herschel, Outlines, p. 654. * See antCy p. 219. 


star a little to the north, from which, as Herschel says, ' a 
foarth nebuloas mass spreads like a fan or downy plume.' So 
far does it spread that Sir John Herschel ^ surmised a con- 
nection between it and another great nebula (M 8), which, 
since it is composed of * nebulous folds and masses, surrounding 
and including a number of oval dark vacancies/ might be 
designated the ' Lagoon ' nebula. Its intermixture with a fine 
cluster has been already mentioned.* 

The spectrum of the trifid nebula consists mainly of the 
inevitable ray at 5005, with a trace of the hydrogen F ; • and 
its light impresses the eye with a corresponding effect of 
greenish colour. Photographic investigations of this object 
can be pursued with advantage only in more southern latitudes 
than ours. 

No logical distinction can be established between irregular 
nebulsB and those indefinite tracts of milky radiance termed 
by the elder Herschel * diffused nebulosities.' The total area 
of fifty-two separate tracts perceived to be thus phosphorescent 
was estimated by him at 152 square degrees, and he added 
the judicious remark that * the abundance of nebulous matter 
diffused through such an expansion of the heavens must 
exceed all imagination.' ^ From the few observations since 
made of these regions,^ it would appear that their illumination 
is not absolutely uniform, cumuli of very sHghtly enhanced 
brightness hinting here and there at initiated condensations, 
and at dim beginnings of nicely-balanced trains of operations, 
set in motion by Sovereign Decree to evoke a cosmos from a 
chaos in those misty fields of space. 

* Loc, cit, * Seo ante, p. 249. ' Secchi, Stilla grande Nehulosa di Orione, 
p. 28. * Phil. Trans, vol. ci. p. 277. 

* Dreyer, V, J, 8. Aatr, Oes, Jahrg. xxii. p. 62 ; Littrow, Stemgruppen und 
Nebelmassen, p. 29 ; Tempel, Astr, Nach, No. 2511. 




Speculations as to an identity of nature between nebulaa 
and comets are no novelty; they presented themselves, as 
they could hardly fail to do, to the mind of Sir William 
Herschel ; ^ but some consistency was first given to them by the 
recent experimental researches of Mr. Lockyer. It is true 
that the results of light analysis are far from being decisive 
in their favour. The spectra of the two classes of bodies are 
fundamentally unlike. No gaseous nebula gives a trace of the 
carbon-bands which characterise ' nearly all comets ; and no 
comet has yet furnished any direct evidence of the presence of 
hydrogen among its constituents. Moreover, nebulae (apart 
from the stars contained in them) seem to emit no genuinely 
continuous light, while cometary nuclei glow in the ordinary 
manner of white-hot solid and liquid substances. Traces of a 
spectroscopic analogy can indeed be shown to exist ; * but they 
are met with only in the secondary elements of each spectrum. 
The resemblance seems only incidental; the dissimilarity 

This does not, however, detract from the closeness of a 
physical analogy, the deep import of which cannot be too 
forcibly dwelt upon. Both comets and nebute consist of 
enormous volumes of gaseous material, controlled by nuclear 
condensations, whether of the same or of a different nature in 
the two genera we need not now stop to inquire. Both, there 
is the strongest reason to believe, shine through the effects of 
electrical excitement. In both, there are manifest signs of the 
working of repulsive, as well as of attractive agencies. 

A telescopic comet is indistinguishable except by its motion 
from the ordinary, centrally brightening globular nebula, which 

> PhU, Trans, vol. ci. p. 306. « See ante p. 76. 



is itself indistinguishable from an exceedingly remote globular 
cluster. Superficial likenesses, however, do not count for much ; 
one object may counterfeit another without bearing any true 
relationship to it. What it is really important to note is the 
structural resemblance of nebulte to comets. The parts of a 
comet become differentiated exclusively under solar influence. 
Hence their symmetrical arrangement, as regards an axis 
passing through the sun, is modified only by the orbital dis- 

Fio. 45. -Nebula in Leo (M 66) 
photographed by M. von Oothard. 

photographed by M. von Oothard. 

placement of the body to which they belong, and from which 
they emanate. It is then extremely curious to find a corre- 
sponding kind of polarity impressed upon the features of certain 
nebulflB. One in Leo, for instance (M 66), photographed at 
Hereny April 16, 1888 (fig. 45), might pass for the head of a 
very* active comet. A series of incomplete envelopes, lying in 
the precisely opposite direction from the nucleus to that taken 
by a bright ray, seem the remnants of successive luminous 
eflfusions under the prescriptive guidance of some external 
force. Forms very similar showed themselves in the comets 


of 1861 and 1882, and in comet Sawerthal (1888).* A fine 
nebula in Gemini (M 65), also photographed by M. von Gothard, 
displays a different kind of axial symmetry. Four wingsy not 
all equally well developed, are attached to a nucleus, in an 
arrangement like that of the sails of a windmill (fig. 46). 
Two-branched spirals no less plainly betray the action of polar 
forces, and confirmatory examples might be drawn from other 
classes of nebulse. 

From many of these objects, however, as from many 
comets, only a single stream of effusion is manifest ; and we 
then get ^ stars with tails,' which might' well pass for minia- 
tures of the bearded travellers through our constellations. 
These effusions are, in some cases, bounded by straight lines ; 
they are fan-shaped; in others, they are curved, often s6 
strongly that the ' brush ' is bent into a * coil.* Such differ- 
ences may be plausibly associated with the varied conditions 
of axial rotation in the star-like nucleus. Where there is 
none, the issuing matter naturally proceeds straight outward ; 
its curvature depends upon its being left continually further 
behind in the continually widening circles reached by it as it 
ascends from an advancing point of efflux. 

It is easy to see that this process, if carried far enough, will 
result in the production of a ' spiral ' nebula. And there sse 
indications that many of these objects are in fact composed 
of matter expelled — perhaps by almost imperceptible degrees — 
from a slowly rotating nucleus, each branch of the spiral cor- 
responding to a separate, though possibly simultaneous efflux. 
Their convolutions are, however, probably of a 'helical' 
nature, — they follow the lines rather of a corkscrew than of a 
watch-spring. The effect seems to be as if the proper 
or advancing movement of the rotating nucleus had been 
accelerated or retarded through influences from which they 
were exempt, while the formation of the branches was still 
progressing. It is not meant to suggest that this actually 
took place, but only to supply a hint towards a mental picture 
of the phenomena we have to deal with. 

» C. H. F. Peters, Astr. Nach. No. 2550 ; Washington Observations for 1880, 
App. I, ; Sirius, Bd. xxi. p. 165. 


"With very rare exceptions, nebalaB are seen by us, not In 
plan, but in perspective. *The only thing,' Professor Holden 
says,^ ' we really know about the form of a nebula, in general, 
is that it is projected into a certain shape. The problem is to 
find the true curves in space, knowing only the projected 
curves.' With this problem, all but insoluble as it appeared* 
he has found the means of dealing, in at any rate a hopeful 

By persevering trials with helices of wire projected in all 
imaginable varieties of position, a curve was at last found 
which might be called typical. In some one of its innumer- 

. ^ able perspective shapes (a 

f Qj ^**— "^^ ^>^^ ^^^ specimens of which are 

given in fig. 47), it proved 
/^"^ \ I capable of representing with 

/^ ^o/ \ V y ^^si^^^*''^!® accuracy the 

v^ ^^ ^"-^ main outlines of nearly all 

Fio. 47.— Type-Curve of Helical Nebul© the nebulsB claSSed aS Spiral, 
viewed in varioas positioDS (Holden). _ . , ,. ,. 

among which the peculiar 
' Omega ' nebula is included. A strong presumption is thus 
raised that the ' type-curve,' or something like it, is really con- 
formed to in nature, and that this first attempt to ' determine 
the actual situation of the different branches of a nebula 
in space of three dimensions from the data afforded by the 
projection of these branches upon the background of the sky/ 
has met with a certain measure of success. 

Should this presumption be borne out by further experience, 
'many interesting questions,' Professor Holden concludes,* 
' may receive a solution. For example, what are the directions 
in space of the axes of these different nebulae ? Is there 
anything systematic in these directions ? What is the law of 
the force by which particles of matter are expelled from (or 
attracted to?) the central nucleus? Have we here in the 
nebulae different types of spirals somewhat analogous to the 
different types of comets' tails so ably discussed by. Professor 
Bredichin? Some of the parts of these nebulae must be 
approaching the earth, some receding from it. Can we by 

' PublicatioTis Astr, Soc, of the Pacific, No. 3. p. 26. * Loc. cit. p. 31. 


the spectroscope discriminate between such motions ? A sug- 
gestion which holds out even the hope of successfully attack- 
ing such problems * has undoubtedly a very high value. 

Nebulae of the spiral and * brush ' kinds are not the only 
ones exhibiting cometary relationships. A characteristic 
feature of nebulous trains and wings, exemplified both in the 
Orion and Pleiades formations, is their vague diffusion on 
one side, but sharp termination on the other, just as in the 
wing-like appendages to the heads of comets. The continued 
ejection of matter against a counter-current of force, by which 
it is unceasingly driven backward, seems indicated in both 

f^ Planetary nebulae, too, imitate, in a fashion of their own, 
the heads of comets under energetic solar action. Their 
multiple discs correspond most strikingly to the multiple 
envelopes of comets, and intimate a similar origin through 
interior expansive or repulsive agencies. Only that, in the 
absence of the dii-ective power of the sun, the waves of ema- 
nation spread equally in all directions, producing successive 
approximately globular, instead of parabolic surfaces. Under 
the combined influenjes of rotation and contraction, such 
shells might be expected eventually to subside into rings ; * but 
it would be extremely rash to affirm that annular nebulae did, 
in point of fact, come in this way to exist. 

Cometary transformations may even help to explain those 
of * new stars.' On January 23, 1835, Halley's comet showed 
as in no respect different from a star of the sixth magnitude ; 
by the 25th it had distended to twice the linear dimensions of 
the planet Jupiter. And Pons's comet, observed September 23, 
1883, as almost stellar, and eight times brighter than the 
day before, had resumed, forty-eight hours later, its habitual 
foggy aspect. The alternate appearances of Nova Cygni as 
star and nebula were perhaps due to causes not unrelated to 
those here concerned. In ascribing to them an electrical 
nature, we risk, it is true, laying ourselves open to the taunt 
ohscunis per obscurior ; yet we can only speak as we know, 
and trust to time and research for better counsel. Elec- 

» lioche, Memoircs de VA'cad, des Sciences, Montpellier, t. viii. p. 244. 



tricity is not the less a potent and universal reality that we 
are ignorant of its mode of working. 

The nebulae are the only class of heavenly bodies from 
which no signs of movement have, up to this, been derived. 
To all appearance, they are absolutely stationary. Accurate 
observations of them are, indeed, comparatively recent ; they 
go back, nevertheless, far enough to justify the statement that 
not one among about four hundred well-determined nebulfe 
becomes progressively displaced by so much as one second of 
arc yearly. The nebulae in the Pleiades ought perhaps to rank 
as exceptions to this general immobility, since we need no 
direct proof to assure us that they drift with the cluster of 
which they form an integral part. The drift, to be sure, is 
slow ; but it is securely ascertained, and aflfords grounds for 
the sole estimate, that is not mere guess-work, of the distance 
of a nebulous system.* The apparent indifference of all other 
nebulae to the perspective effects of the sun's swift advance 
through space, leaves little probabiHty that any of them lie 
as near to us as the nearer Stars ; and there are indications 
besides that whatever individual movements they possess are 
of an extremely sluggish nature. For none yet experimented 
upon display, to a measurable extent, the spectroscopic line- 
displacements indicating motion in line of sight ; and since 
these have no relation to remoteness or. vicinity, the inference 
is obviously suggested that nebulae are really slower moving 
bodies than stars. Decisive and suflScient information on this 
point cannot, however, be looked for until photography has 
been brought to bear upon it. 

There is the same total want of evidence of orbital, as of 
translatory movements in nebulae. The systems they in 
numerous cases presumably constitute remain rigidly fixed. 
The contrary has often been asserted ; yet revolutions, alleged 
on the strength of inexact observations, and brought to a 
standstill by precise ones, must plainly be dismissed as illu- 
sory; 'unless,' as Dr. Dreyer says,* 'we are to believe that 
nebulae in the good old days moved about as they liked, but 
have been on their good behaviour since 1861 and kept quiet.' 

* See ante, p. 226. « Monthly Notices^ vol. xlvii. p. 418. 


The existence, so frequently observed, of a nebulous connection 
between grouped nebulous objects intimates a state of things 
hardly reconcilable with mutual circulation. The relations of 
these, as yet, imperfectly separated individuals are perhaps in 
a state of transition, like those of multiple cometary masses, 
at times enclosed together, like double nebulse, in a dimly 
luminous sheath. 

The idea of the accompaniment of planetary nebulsB by 
satellites was suggested to Sir John Herschel by the frequent 
and close proximity to such objects of minute stars,* and he 
recommended their careful micrometrical measurement as a 
criterion of possible future changes. But none have yet been 
detected. D'Arrest found the attendant stars just in the 
positions Herschel had assigned to them,^ and they have not 
visibly shifted since. The probability, indeed, of the stars 
genuine association with the nebulae they seem to wait upon 
has gained in strength with the general increase of knowledge 
regarding stellar and nebular relations ; nevertheless, circu- 
latory movements, if in progress, are likely (considering the 
enormous remoteness and consequent spaciousness of these 
systems) to be so exceedingly slow that centuries are but as 
days in the reckoning of their periods. 

Systemic alterations in nebulaB may then long evade re- 
cognition, and their geometrical status quo long be apparently 
preserved. Possibly, however, an exception to this rule may 
be found in the second companion to the great Andromeda 
nebula, should the pivoting movement suspected to have 
been executed by it between 1876 and 1888 be verified by 
continuance. Otherwise, the assertion of recent photographs 
regarding the direction of the longer axis of the attendant 
nebula must be considered to hold good for every anterior 
epoch. Changes arrested by the exclusion of the possibility 
of mistakes may pretty safely be set down as having originated 
in them. 

Variability in light is a quality of nebulsB as surely as of 
stars, although one, in particular cases, by no means easy 
to establish. Nebula are peculiarly sensitive to atmospheric 

' Phil. Trans, vol. cxxiii. p. 500. * Leipzig AbJiandlungen, Bd. iii. p. 308. 

c 2 


influences. Their finer details, always hovering on the verge 
of visibility, are completely shrouded by the lightest mist. 
Hence, even to the same eye, and with the same instrument, 
the aspect of the same nebula often varies greatly from night 
to night ; and since personality is nowhere stronger than in 
the perception of the delicate luminous gradations delineating 
to our sight the forms of nebulie, a difference of observers 
adds a further incalculable element of uncertainty. Eumours 
of change then easily arise, but are with diflSculty sub- 

The presumption they start from is, however, fully war- 
ranted by facts. The occurrence of luminous fluctuations 
in nebuleB has been proved by the total disappearance of three, 
all, strange to say, situated in the same constellation. On 
October 11, 1852, Mr. Hind discovered, near the group of the 
Hyades in Taurus, a small round nebula (N. G. C, 1555) with 
slight central condensation. It was then very faint, but 
brightened steadily until 1856, when d' Arrest ranked it as 
belonging to the first, although verging towards the second 
class of brilliancy.* His amazement then was extreme to 
find on October 3, 1861, its place apparently vacant ! Some 
glimmer of its light was indeed made out for a year or two 
longer with the Pulkowa fifteen-inch refractor ; but that too 
faded, and the object has now, for above a quarter of a century, 
been invisible in the most powerful telescopes. The appari- 
tion, so far as can be judged, was a strictly temporary one. 
Hind's notice, there is reason to believe, did not lag far 
behind its first perceptibility with moderate instruments. 

A curious feature of the occurrence was the sympathetic, 
or at any rate simultaneous, decay in light of a small star — 
since known as * T Tauri ' — placed almost in contact with the 
nebula. The star, however, recovered in 1868 about the same 
time that a second new nebula (N.G.C., 1554) came into view. 
First discerned by 0. Struve, it was observed by d'Arrest, w^ho 
was fully convinced of its novelty ; and his opinion was borne 
out by its subsequent total disappearance.^ 

» Astr. Nach. Nob. 13G6, 1080 ; Auwers, ibid. No. 1391. 
' Dreyer, Matioirs R. Astr. Soc. vol. xlix. p. 214. 


The third ' temporary ' nebula within astronomical ac- 
quaintance was seen by Chacornac at Paris, attached to an 
eleventh magnitude star near 5'Tauri, October 19, 1855 (N. G. C, 
1988). A conspicuous gain in lustre allowed it to be perceived, 
after three months, as covering a nearly rectangular area, 
^V ^7 ^¥f ^'ith thin parallel strise, like those of a cirro- 
stratus cloud.^ On November 20, 1862, it was nevertheless 
utterly gone, the star remaining unafifected alike by its 
visibility and by its disappearance. 

The light-changes of nebulae do not offer the same diversity 
as those of stars. Only two kinds of variabiUty — those 
producing respectively ephemeral appearances and capricious 
brightenings and fadings — are represented among them. No 
periodical nebulee have yet been shown to exist. The influences, 
of whatever nature, bringing about the rhythmical pulsations 
of stellar light would seem to be absent from the nebular 
kingdom, A distinction, however, peculiar to themselves, can 
be estabhshed among variable nebulae. Their fluctuations 
may be either general or partial. They may affect the whole 
of a moderately compact object, or certain sections of an 
extensive formation. Examples of both kinds, and of all 
degrees of authenticity, abound ; but we will only mention a 
very few, in which the reaUty of change seems scarcely 

One such is afforded by an elliptical nebula in Leo (N. G. C, 
8666), * very bright ' when discovered by the elder Herschel 
in 1784, but noted by his son as abnormally faint for the 
first class. Subsequently observed alternations have made it 
all but certain that the discrepancy indicated genuine change.^ 
A nebula in Cetus, too (N. G. C, 955), is evidently subject to 
similar vicissitudes. Schonfeld in 1861, and Vogel in 1865, 
failed to see it, although it was, at sundry other epochs, easily 
visible to the former observer, as well as to d' Arrest and 
Winnecke, and fully justified in 1887, in Dr. Dreyer's opinion, 

* Cojnptes Rendus, t. Ivi. p. 637. 

* Winnecke, Astr, Nach. No. 2293 ; Dreyer, Memoirs R. Astr. Soc. vol. xlix. 
p. 218. 


Herschers classification of it as of the second order of bright- 


Again, one of a group of nebulae in Virgo, observed by 
Schmidt at Athens in 1862 (N.G.C., 5655), could not be found 
by d' Arrest in 1865, two minute stars appearing as its locum 
tenentea.^ If, as seems probable, the identical object was 
recorded by Herschel, from an observation of December 28, 
1785, as No. 498 of his second class, its re-emergence to view 
may at any time be looked for. The collection of objects 
with which it is associated were judged by d' Arrest (no doubt 
rightly) to be the brightest * knots ' of a wide-spreading 
nebulous structure (M 49). The variability of one of them 
approximates then to the local changes of irregular nebulae, 
exemplified with most certainty in their chief, the great Orion 

The occurrence of such in this well-watched object has been 
placed beyond reasonable doubt by the researches of Professors 
0. Struve ' and Holden.* Even while the observations of the 
latter were in progress, fluctuations in the lustre of some of the 
component masses were distinctly recognisable, and one new 
nebulous patch was actually seen to develop out of a faint 
stellar point. No measurable change of place, however, could 
be detected by a most diligent comparison of the records of 
over a century, in either the misty floccules or the contained 
stars. Some of these, of the minuter sort, are certainly 
variable to the extent of one or two magnitudes. And their 
changes, although apparently lawless, are at times very rapid, 
as was evidenced by the striking inequalities of the coupled 
images of ten or twelve of them on a plate doubly exposed by 
Mr. Roberts, at an interval of four days, early in 1889. 

The partial light fluctuations, strongly suspected in both 
the Omega and the Trifid nebulae, have resulted chiefly in a 
modification of what we may call coast-lines, here advancing, 
there encroached upon by the sea of darkness which surrounds 

" Winneoke, Monthly Notices, vol. zzzviii. p. 104 ; Dreyer, loc. cit, p. 213. 
« D'Arrest, Asir. Nach. No. 1520. 

• F. J. S. Astr. Ges. Jahrg. xix. p. 35 ; Melanges Math. t. ii. p. 530 ; d'Arrest, 
Astr. Nach. No. 1366. 

* Washington Observaiums, 1878, App. i. pp. 121, 225. 


them. In the latter object, a singular apparent alteration in 
the relative places of the multiple star and the nebulous 
masses involving it, is perhaps due to this instability of out- 
lines. Sir John Herschel in 1827 and 1833 described the 
star as located * exactly in the central vacuity ' of the nebula, 
and just at the point of convergence of the three rifts dividing 
it throughout.* But a drawing made by him at the Cape in 
August 1835 exhibits the star no longer as central, but as 
adhering to the eastern mass of nebulosity ; and virtually the 
same state of things was noted by Mason and Smith in 1839,* 
and subsists so obviously to the present time as to render a 
mistake about it inconceivable. The implied change, however, 
must have taken place abruptly between 1833 and 1835, and 
then ceased ; so that proper motion, either of the star or 
nebula, had certainly nothing to do with it.' There seems 
then no alternative but to admit that the frontier-lines 
between luminosity and obscurity were, at the epoch in ques- 
tion, very considerably ' rectified.* 

The relative variability of the parts of nebulae will hence- 
forth, in all probabiUty, be investigated l?y photographic means 
alone. Data of perfect definiteness on the subject can now 
easily be obtained by the use of a light-scale previously 
impressed upon the plates to be exposed ; and their accumula- 
tion during some years will do much to abolish the doubts 
that still remain as to the extent and reality of the shiftings 
of the luminous level in these still enigmatical objects. 

One point about them appears, however, tolerably clear. 
It is that the glow of those shining with discontinuous light 
is of an electrical nature.* For the gases of which they are 
mainly composed cannot be rendered incandescent by the 
transference of heat from diffused solid particles or bodies 
not themselves highly incandescent; and the absence from 

* Memcdrs B. Astr. 8oc, vol. iii. p. 63 ; Phil. Trans, vol. cxxiii. p. 460 ; 
Holden, Amer. Jour, of Science^ vol. xiv. p. 434 (1877). 

* Trans. Amer. Phil. Soc. vol. vii. p. 176. 

■ Dreyer, Monthly Notices^ vol. xlvii. p. 419. 

* In his Bakerian Lecture, Mr. Lockyer stated this to be 'proved' as 
regards the glow of hydrogen in the Orion nebula {Proc. R. Soc. vol. xliv. p. 
12), and his argument is valid for all gaseous nebulee. 


nebulae of glowing solids is shown by the nature of the nebular 
spectrum, which probably includes no element of truly con- 
tinuous light, apart from that of the * stars,' apparently 
associated in some way with all nebulae. The unmistakable 
analogy, again, between solar-coronal and cometary forms on 
the one side, and nebular forms on the other, indicates for all 
a kindred origin in the play of opposing forces, generated by 
certain foci of condensation, one of which is our sun, while the 
others may be vaguely but safely designated as * nuclei.' Where 
there is only one such nucleus, the enveloping gases assume 
a simple globular or oval shape ; where there are many,, the 
result is exceedingly complex. Irregular nebulae are thus, in 
our view, potential star-clusters; they consist of a stellar 
framework, draped with nebulous folds, spirals, and festoons, 
disposed along lines of force laid down by the rival or con- 
current energies of the compact masses which it is permissible 
to regard as inchoate suns. 




The most arduous among the problems of stellar astronomy 
was, singularly enough, the first to be attacked. It was 
attacked, indeed, before the possibility was even remotely dis- 
cerned that stellar astronomy might come to be regarded as a 
substantive branch of science. In the hope, not of penetrating 
the inscrutable secrets of the remote sphere of the fixed stars, 
but of solving doubts about the motion of the earth, Copernicus, 
Tycho, and Galileo led the way in the long series of experi- 
ments on the apparent displacements of the stars resulting 
from our own annual travels round the sun. The interest of 
the question whether such displacements existed or not was 
for them of a wholly * parochial ' kind ; it lay in the test they 
afforded as to the reality of the terrestrial revolutions. Should 
the stars be found to shift ever so little by the effect of per- 
spective, then the heliocentric theory could no longer be gain- 
said; if, on the contrary, they ignored sublunary circlings, 
the ' pill ' (as Kepler termed it) to be swallowed by Copernicans 
was indeed a huge one. For the distances to which the fixed 
stars had, in that case, to be relegated, seemed in those times 
monstrous and incredible ; and monstrous and incredible they 
would appear still, were we not forced by irrecusable evidence 
to believe in them. 

From the beginning to the end (so far) of the history of 
these inquiries, it may be taken almost as an axiom that the 
largest ostensible parallaxes have been obtained by the worst 
means. With each successive improvement in methods and 
instruments, as the limits of possible error shrank, the dis- 
placements apparently measured dwindled, and the stars 
became less accessible to attempted determinations. During 


Bome three centuries, the ill-success in this matter of an 
astronomer was a measure of his skill and judgment. Besults 
obtained with suspicious facility by inexpert observers utterly 
evaded the guarded scrutiny of such men as Tycho Brahe, 
Bradley, and Pond. Flamsteed, indeed, just at the close of 
the seventeenth century, detected in the pole-star annual varia- 
tions which were certainly not illusory. Yet here too there 
was a caveat. Theory and fact did not correspond. 

Let us consider for a moment what must be the visual 
effects upon very distant objects of the comprehensive and 
unceasing rounds of the planet upon which we are borne as 
spectators. Unmistakably, to begin with, we see them in 
different directions at different times of year. In January and 
July, in March and September, and so on, we are at opposite 
ends of base-lines 186 millions of miles in length. The stars 
then must be continually thrown, now a little to one side, now 
to the other, of the true, or * mean ' places which they would 
severally occupy if viewed from the immobile sun. In other 
words, each describes round its mean place in a period of a year 
a small apparent orbit, which is nothing else than the orbit of 
the earth projected in miniature on the sky. For stars situated 
in the ecliptic — that is, in the plane of the earth's travelling — 
this orbit contracts into a right line, along which the star merely 
swings to and fro ; for stars near the pole of the ecliptic, the per- 
spective orbit is virtually a circle ; while intermediate latitudes 
afford all degrees of foreshortening. Every star — unless those 
few lying close to the pole of the ecliptic — ^has thus its epochs 
of maximum parallax, six months apart, when it seems to stand 
alternately at opposite extremities of the major axis of the par- 
allactic elUpse, and it is then that measures of its apparent 
displacements can be most advantageously made. These 
opportune seasons occur when the earth's longitude faUs short 
of, or exceeds by ninety degrees the longitude of the star. They 
are accordingly different for stars with different longitudes. 

,The precise /orm of displacement due to the earth's revolu- 
tion round the sun is thus strictly calculable for each individual 
star ; the amount alone cannot be predicted, but must be obtained 
by observation ; and from this amount the distance of the star 


is deduced. For each parallactic orbit is a perfect model, both 
in shape and size, of the earth's orbit as it would be seen from 
the star, abridgment of compass (down to contraction into a 
virtual point) corresponding to a more and more profound im- 
mersion of the point of survey in the abysses of space. 

The parallax of a star is then the difference between its 
positions as seen from either side, and from the centre of the 
earth's orbit. It is, in short, the angle subtended, at the dis- 
tance of that particular star, by the mean interval between 
earth and sun. Now we can tell in a moment how far off a 
spectator must be to see a line ninety-three millions of miles 
in length diminished to the angular dimension of, let us say, 
one second. He must be 206,265 x 98 millions of miles 
distant. But no star has yet been found so near to us as this. 
That is to say, the shift of no known star amomits to as much 
as the width of a sixpence held up at Charing Cross to a spec- 
tator at Stanhope Gate or at Millbank. 

We are now in a position to understand why it was that 
Flamsteed's observations of the apparent displacements of 
Polaris could not, when critically examined, be set down to the 
account of parallax. The star seemed indeed to describe, 
regularly each year, a little ellipse of exactly the right shape ; 
and as to its size, there was no a priori reason why the pole- 
star should not have a parallax of upwards of twenty seconds. 
But there was one irreconcilable discrepancy. The displace- 
ments noted occurred at wrong times. Had they been of a 
parallactic nature, the position of the star in its minute ficti- 
tious orbit should have been invariably ninety degrees in 
advance of what it actually was. They were not then due to 
parallax ; but obtained their proper explanation from Bradley's 
discovery of the aberration of Ught in 1729. 

During the ensuing century and a quarter, the only valid 
results obtained in this direction consisted in demonstrations^ 
renewed and enforced from time to time as more conclusive 
evidence presented itself, that with the instrumental means 
then available, stellar parallax was an inappreciable quantity. 
Bradley showed that it must fall short of half a second, and 
although his reasoning applied strictly to only a limited number 


of stars, it wae ren lered at once more general and more cogent 
by the investigations of Pond and Airy, of Struve and Bessel. 
It thus seemed that astronomers should content themselves 
with the knowledge that the stars were exorbitantly remote— so 
remote that light spent at least four or five years in travelUng 
to us from the brightest of them, and might, for anything that 
appeared, spend indefinitely longer. The labours and refine- 
ments of two centuries had issued in fixing a lower limit for 
distance, an vjqier limit for jmrallax ; in isolating the sun from 
his compeers by setting between him and them an unmeasured 
abyss of desert space ; in widening to a startling extent the 
boundaries of the visible universe. Kepler's * mighty bolus 
had to be swallowed in its entirety. 

At last, in 1827, Savary of Paris brought forward a method 
(already referred to * ) for fixing an up2}er limit to the distances, 
a lower to the parallaxes of binary stars moving in known 
orbits. The further off from us such orbits are, the greater of 
course their real size, and the longer the time taken by light 
to cross them. Hence, the deviations from their true places 
of the moving stars caused by the delay in our vision of one of 
them, serve in theory, since they increase with remoteness, to 
determine the distance from the earth of the pair. Or if no 
such deviations are apparent, it should at least be possible 
to fix an amount which they could not exceed without becom- 
ing so. Savary, accordingly, professed to demonstr'ate in this 
way that f UrssB Majoris, the couple most favourably situated 
for the purposes of the inquiry, must have a parallax exceeding 
3 ^ of a second ^ — must, that is, be at a less distance than 
would be traversed by light in 109 years. But the informa- 
tion, however credible in itself, was not fully authenticated. 
Villarceau showed in 1878 ^ that the method was inapplicable 
except to stars of known relative masses ; and these as yet 
scarcely exist. The most that can be hoped for from Savary's 
light-inequality is that eventually, in some rare cases, it may 
serve to determine indirectly parallaxes too small for direct 
measurement. Among the few pairs besides f Ursa) in. 

• See anU, p. 202. 

2 Con7i. des Temps, 1830, p. 169. ^ IbUl. 1878, p. 68. 


which it might be perceptible are f Herculis and rf Coronas 

By-ways, however, are of secondary importance since the 
straight road to the end in view has been made practicable. 
The engineer who carried it over the barrier long helplessly 
confronted was Fraunhofer. The improvements eiBfected by 
him in the power and perfection of telescopes, as well as in the 
application to them of micrometrical apparatus, alone rendered 
possible the exquisitely refined measurements upon which the 
detection of stellar parallax absolutely depends.* 

These measurements are for the purpose of determining 
variations in the angle between the star chosen as the subject 
of experiment and one or more * comparison-stars ' in its 
neighbourhood, treated as fixed points of reference because 
assumed to be indefinitely remote. From progressive annual 
changes of the intervals separating them from the star they 
serve to test, the amount of its parallax, hence of its distance 
from the earth, is learned. But if the comparison-stars are 
themselves afifected by sensible parallactic displacements, then 
the result is, to a certain extent, if not wholly, vitiated. Should 
they chance, for instance, to be at the same distance from us 
as the compared object, then all the stars under observation 
will shift together, giving the effect of immobility, and imply- 
ing the absence of measurable parallax, when in reality a 
large one may be present. Again, a parallax sought by the 
aid of a comparison-star itself possessing one-half the parallax 
of the star investigated, will come out only one-half of its true 
value. It may even happen that the comparison-star is, of 
the two, our nearer neighbour in space, when a 'negative 
parallax * (as it is called) will emerge, showing how great is the 
discrepancy in the wrong direction. 

It is, indeed, the weakness of the ' differential method of 
parallaxes ' that it gives relative, never absolute results. Not 
only does some degree of doubt always attach to them, but 
their deviations from the truth are always on the same side. 
They tend in all cases to diminish the parallax, and exagge- 

' Birkenmajer, Sitzungsberichte, Wien, Bd. xciii. ii. p. 738. 

^ C. A. F. Peters, Ztitschrift fib' populcire Mittheilungeny Bd. iii. p. 96. 


rate the concluded distance. Nevertheless, the advantages of 
the method are so overwhelming, it abolishes so many causes 
of error, and strikes at the root of so many illusions, that it 
has gained universal and exclusive preference. No one any 
longer thinks of attacking this delicate problem by the com- 
paratively clumsy mode of determining absolute right ascen- 
sions and declinations at intervals of six months, and trying 
to distil, from a confused mass of pervading infinitesimal 
errors, the all but insensible evidence of those periodical 
changes which reflect across the gulf of sidereal space the 
movement of the earth in its orbit. 

Nevertheless, the first genuinely measured stellar parallax 
was so far a casual result that it was arrived at in the ordinary 
course of observation. It presented itself, as it were, un- 
solicited. Alpha Centauri combines Struve's three criteria of 
vicinity. It is exceedingly bright; it has a large proper 
motion ; and its components revolve swiftly in a wide orbit. 
No other star in the sky seemed beforehand so likely to come 
within the grasp of terrestrial determinations ; and none has 
yet proved to be so little remote. Henderson's observations 
of it, made at the Cape in 1832-3, were discussed with a view 
to sifting out from them a parallactic element, after he had 
learned that the star's rapid onward movement afforded a 
presumption of relative nearness to the earth. Nor were his 
expectations beUed. A parallax of about one second mani- 
festly implied by annual changes of declination, was partially 
confirmed by observations of the same object in right ascen- 
sion made by Henderson's assistant. Lieutenant Meadows. 

His announcement to this effect, January 3, 1839,* was 
however, received with doubts, justifiable perhaps, consider- 
ing the numerous precedents for illusion on the point, but not 
justified by the event. Bessel's similar (and slightly prior) com- 
munication regarding 61 Cygni inspired, on the other hand, 
general confidence. The Konigsberg series, indeed, though 
by no means fortified with all the precautions now deemed 
necessary, seemed beautifully complete. The observations 
were of the differential kind, and favourable opportunities for 

' Memoirs li. Astr, Sor. vols. xi. p. 61, xii. p. 32ri. 


measurement were afforded at intervals of three months by 
the use of two comparison-stars so situated that the lines of 
maximum displacement with respect to them of 61 Cygni lay 
at right angles to each other. The close accordance in their 
results of operations thus wholly independent, and executed 
under widely different influences of season and temperature, 
and the harmonious flow of the curves into which the obser- 
vations were projected by Mr. Main,^ prompted the conviction 
that here at last was a stellar parallax the genuineness of 
which was beyond cavil. The extreme importance of its 
detection, pronounced by Sir John Herschel the greatest 
triumph ever achieved by practical astronomy, can be esti- 
mated from Bessel's declaration that, until it was actually 
compassed, he was unable to form an opinion as to whether 
the parallaxes of the nearest stars should be reckoned by 
tenths or by thousandths of a second ! ^ 

The distance from the earth of 61 Cygni has been more 
frequently investigated than that of any other star, and, some 
trifling discrepancies notwithstanding, may be considered as 
satisfactorily ascertained. Bessel's parallax of about a third 
of a second turned out to be slightly too small, but was aug- 
mentedto 0''-42 by Auwers'srediscussion in 1868 of the same 
data.^ The two most reliable determinations are, perhaps. 
Sir Eobert Ball's in 1879, giving a value of 0''-465, and 
Professor Pritchard's, giving 0"-432 from photographic com- 
parisons in 1887. On the whole, the parallax of this star can 
hardly differ much from 0"-45. 

Although the third largest known, it yet implies an actual 
distance so inconceivably vast that light spends seven years 
and three months in flying over the forty millions of millions 
of miles serving to measure it. Thus we see the coupled 
stars, not where they are, but where they were seven and a 
quarter years ago ; that is (since their proper motion is about 
5''-14 yearly) just thirty-eight seconds of arc behind their true 
places. The effulgent points terrestrially determined are 
then mere simulacra of the real stars ; they pursue, without 
ever overtaking them ; they would continue to shine and to 

» Ibid. vol. xii. p. 42. ■ Astr. Nach. No. 385. 

» AbJia7idlungen KiM. Akad. Berlin, 1868. p. 114. 


travel for seven years and upwards after their originals had 
been blotted out of the visible creation. Our views of all 
moving objects are of course to some extent affected by this 
curious kind of light-aberration ; but in the sidereal heavens 
it attains proportions that are not only large, but, for the 
most part, incalculably large. Our survey of the background 
of the sky may lag centuries, even millenniums, behind our 
simultaneous survey of its foreground ; and the disturbed 
synchronous relations between the varied luminous contents 
of the sphere are, so far as our perception goes, unsusceptible 
of reconstruction. 

Transported to the place of 61 Cygni, our sun would 
appear more than eight times as bright. It would represent, 
not a star below the fifth, but one above the third magnitude, 
such as B CassiopeifiB, one of the lucid five constituting the 
* W ' by which that constellation strikes the eye. The Swan 
binary, indeed, ranks at present (for it may have seen better 
days) among the least luminous of the stellar host. 

Fraunhofer's construction of the instrument with which 
Bessel observed it marked the turning-point from failure to 
success m parallactic inquiries. The heliometer, as a star- 
measuring machine, not through superimposed contrivance, 
but by original design,* is specially adapted to facilitate them, 
and may almost be said to have, in this department, super- 
seded the ordinary equatoreal and micrometer. It has two 
chief points of superiority. In the first place, much wider 
paii'S of stars can be grasped with it, its compass being, by the 
mobility of the semi-lenses, extended far beyond the limits of a 
single field of view. The range of selection for comparison- 
stars is thus greatly enlarged, and the chances of a systemic 
connection with the central star fatal to the purpose of the 
designed operation are reduced to a minimum. In the next 
place, the stars under observation can be visually equalised 
by placing a wire-gauze screen of any desirable opacity over 
the segment of the object-glass forming the image of the 
brighter one, whereby tendencies to personal error, difficult 
to be otherwise got rid of, are perfectly eliminated. These 

* See aritt', p. 15. 


are not the only features of the heliometer tending to promote 
critical precision, but they are the most important. 

Bessel's success with 61 Cygni gave the impulse to 
numerous undertakings of the same kind. Their result 
depended mainly on the skill or luck of the observers in 
picking out from innumerable, indefinitely remote stars the 
few near enough to be sensibly displaced by being viewed 
from opposite sides of the earth's orbit. Two circumstances 
mainly determined their choice. 

That distance is a factor of stellar brightness is so obvious 
a truth that it may almost be reckoned a truism. Admitting 
the widest possible range of variety in actual light-power, it 
is still certain that the most lustrous objects are likely to 
be found in closest proximity to the earth. That is, on a 
wide average. Individual exceptions abound ; but the more 
numerous the stars considered, the more approximately does 
the theoretical inverse ratio between distance and the square 
root of total light correspond to actual fact.^ 

Several conspicuous stars were, on this ground, fixed upon 
for investigation by C. A. F. Peters in 1842-3. His instru- 
ment, the Pulkowa vertical circle, was perfect of its kind, and 
was used with consummate skill. The results, moreover, so 
far as they went, were absolute, being irrespective of com- 
parisons with other stars ; but they did not go far. The 
deduced parallaxes were so small, and their ' probable errors ' 
so relatively large, that it was difficult to place much confidence 
in them. Vega came out with a parallax of one-tenth of a 
second, Arcturus with O'^'IS, each uncertain to the extent of 
more than half their respective values ; the possible displace- 
ments of Capella were completely masked by observational 
discrepancies ; a Cygni remained rigid, the measures only 
serving to show four hundred chances to one against its 
possessing a parallax exceeding O'^'l.^ Only the result for 
Polaris was, owing to the exceptionally advantageous position 

' Struve, Mens. Microm, p. clxii. 

' Zeitschrift fUr pop. Mitth. Bd. iii. p. 105 ; M^moireSt St. P^tersbourg, 
t. vii. p. 140, 1853 ; Astr. Nach. No. 1147. 


of the star, thoroughly trustworthy, and has since been 

A second criterion of nearness was found in the appearance 
of rapid motion. This varies in the same proportion as 
distance, but in the reverse sense. At twice the distance, 
an identical velocity is only half as effective in producing 
angular displacement ; at three times the distance, its seeming 
amount is reduced to one-third, and so on. Thus, apparent 
swiftness, no less than apparent lustre, depends in part upon 
vicinity, and the largest proper motions must l)elong, on the 
whole, to the nearest stars. 

And, on the whole, parallax-hunters, taking rapidity of 
advance for their guide, have prospered the best. A seventh- 
magnitude star in Ursa Major, flitting annually over 4| seconds 
of angular space, was found by Winnecke in 1858 ' to have a 
parallax (=0"'5) inferior only, among those as yet determined, 
to that of a Centauri. This insignificant object, numbered 
21,185 in Lalande's great catalogue, is separated from the 
earth by a light-journey of 6^ years, and to that extent our 
observations of it are retarded. So that it is in reaUty always 
81'' in advance of the place we are compelled to assign to it. 
For a body claiming the rank of a sun, it is either very small 
or very obscure. In its position, our ruling orb would show 
as fifty-seven times more luminous. 

An 8'5 magnitude star in the same constellation (Lalande, 
21,258), also distinguished for apparent velocity, disclosed to 
Auwers's measurements with the Konigsberg heliometer in 
1860-2 a parallax of (y'-26,* corresponding to a light-journey 
of 12^ years, and a permanent ^displacement on the sphere, 
due to its proper motion in that interval, of 55''. The real 
brilliancy of the star is only -^ that of the sun. Kriiger's 
simultaneous and identical determination of its distance with the 
Bonn heliometer showed, as he remarked, the peculiar aptitude 
for such researches of that description of instrument."* A 
still smaller star in Draco (Oeltzen, 17,415) gave an even 
more emphatic warrant to confidence in swiftness, rather than 

> Astr. Nach, No. 1147. ' Monthly Notices, vol. xxiii. p. 74. 

* Acta Societal. Scient. Fmnicce, t. vii. p. 373. 


in conspicuousness, as a certificate of proximity. Kriiger, 
induced by its yearly movement of l''-27 to subject it to experi- 
ment, obtained a parallax of one quarter of a second ; ^ while 
the fine binary system, 70 Ophiuchi, with an annual motion 
of 1", proved to be removed from the earth by twenty years 
of light-travel (parallax 0"*16).* And all these results seemed, 
from the smallness of their ' probable errors,' to be exceedingly 

The probable error of any result, however, represents only 
what we may call the uncaused inaccuracies of the observations 
upon which it is founded. It sums up, according to the doc- 
trine of probabilities, the effect of their deviations, in either 
direction, from the mean. But it takes no heed of ' systematic ' 
errors due to causes working steadily in one sense, but, so to 
speak, underground. These are the real sources of mischief 
from which fallacious parallaxes have abundantly sprung in 
times past, and which cannot, in the present and future, be 
too carefully guarded against. Especially formidable are 
certain slight idiosyncrasies of perception, by which measures 
of distance become modified with the varying positions of the 
line of direction between the objects measured, relative either 
to the vertical or to the line joining the observer's eyes. 
And since this subtle spring of error rises and falls har- 
moniously in a period of a year (because dependent upon the 
uranographical situation of the stars under scrutiny), it 
would be capable, if not adverted to, not only of completely 
vitiating observations apparently accordant, but even of simu- 
lating parallactic changes that had no real existence. Instru- 
mental errors, too, connected with changes of temperature, or 
the deforming power of gravity (as conditioned by the shifted 
positions of the telescope at different seasons), take the same 
cyclical course ; and there can be no doubt that to some such 
lurking deceptive influence the parallax of 0""07 attributed to 
a Herculis by Captain Jacob in 1858,' owed its origin. Since 

* Acta Societat Scient. Fmnic<Bt t. vii. p. 383. 

* Kruger, Astr. Nach, Nos. 1210-12. 

■ Madras Observations, 1848-52, Appendix ; Memoirs R, Astr, 8oc. vol. 
zxvii. p. 44. 

X 2 


the star chosen for comparison was no other than the well- 
known physical attendant of the object examined, the fact of 
illusion is patent. 

The exigencies of this kind of work were first recognised, 
and its principles fully explained, in an elaborate paper pub- 
lished by Dr. Dollen of St. Petersburg in 1855 ; ' and his 
principles were ably carried into effect by Dr. Briinnow in a 
series of investigations of stellar parallax at Dublin between 
1868 and 1874. The example thus set of the thorough elimi- 
nation of errors, at once personal and periodical, has since 
been generally followed. More effectually than by most other 
men, the famous ' Know thyself ' of the old Greek philosophers 
has been taken to heart by astronomers. Their anxious and 
elaborate inquiries regard not merely microscopic inequalities 
of scale-divisions and screw-values, changes in refraction, 
corrections for aberration and proper motion, but the cunning 
tricks of their own nerves, the caprices of cerebration, all the 
varying conditions of perception in the organism at their 
individual command. 

None of these precautions were neglected in the important 
work executed by Drs. Gill and Elkin at the Cape in 1881-3.' 
Fully alive to its subtle requirements, they gave to their deter- 
minations a precision which entitles them to rank among the 
most valuable of astronomical data. Dr. Gill's discussion, 
especially, of the parallax of a Centauri is a model of what 
such an inquiry should be. It leaves, one may say, no stone 
unturned beneath which a source of illusion might lie 

Of the nine southern stars undertaken, three were investi- 
gated concurrently by both observers. Among these was 
a Centauri, and the parallax of 0"'75, resulting from independent 
comparisons with no less than four pairs of adjacent stars, is 
probably more nearly accurate than any value of the sort yet 
registered. The fact is then assured that light, which flies 
from the sun hither in eight minutes, spends four years and 
four months on the journey from the nearest fixed star. The 

* Bulletin de VAcad. St. P^tersbourg, t. xiii. Snppl. 
■ Memoirs B. Astr. Soe. vol. xlviii. 


corresponding distance is, in round numbers, twenty-five 
billion miles. 

On Sirius, too, a double attack was made, of which the 
upshot was a parallax of 0"'38, or a distance of 8-6 light- 
years. These were the first measures of the bright dog- 
star made under perfectly suitable conditions. They are at 
present being repeated by Dr. Gill with different comparison- 

The third parallax simultaneously determined was that of 
s Indi, a fifth-magnitude star with a proper motion of nearly 
five seconds a year. It proved to be 0''-22, representing a 
light journey of 14^ years, and showing that the total radiative 
power of the star is about half that of the sun, while its rush 
through space, taking only the directly measurable,, thwart- 
wise part of its motion, is at the rate of sixty-three miles per 
second, or more than three times as swift as the earth's move- 
ment round the sun. 

Of the six remaining stars, four, chosen for their ex- 
ceptional mobility, did not fail in some degree to justify the 
presumption it afforded of relatively near neighbourhood. 
Thus, the chief star of the remarkable triple system, Oj Eridani, 
yielded to Dr. Gill's investigation a parallax of 0''*166, im- 
plying a light-journey of 19-6 years, and a ' proper motion 
displacement ' for the trio of 80". The swiftest moving star, 
however, in the southern hemisphere is one of 7*5 magnitude 
in Piscis Australis, known as * Lacaille 9352.' Its angular 
rate of progress only just falls short of 7" a year ; and the 
parallax of 0"'28 (distance 11^ light-years) obtained for it by 
Dr. Gill proves its linear velocity to be at least 73 miles per 
second. A still quicker traveller is f Toucani, of fourth 
magnitude and circumpolar at the Cape, which, with a proper 
motion of two seconds, is discovered to be advancing at the 
express rate of 101 miles a second. That is, if Dr. Elkin's 
small parallax of O^'-OG, corresponding to a light-journey of fifty- 
four years, be correct. In proportion to its amount, it must be 
admitted that the margin of error qualifying it is somewhat 
unduly large. The last of the rapid stars measured w^as 
€ Eridani, for which Dr. Elkin's operations brought out a 


parallax of 0'''14, representing a light-journey of twenty-two 
years and nine months. This star is of 4^ magnitude, has a 
proper motion of more than 8'', and is about thrice as luminous 
as our sun. 

The two remaining stars on the Cape list, though both 
splendid objects, are virtually fixed in the sky. They have no 
perceptible onward motion ; and neither of them, accordingly, 
showed any sign of possessing a sensible parallax. This is 
really, when we come to consider it, an astonishing result. 

Second only to Sirius in the southern hemisphere, Canopus 
far outshines every star north of the celestial equator. As the 
chief luminary of the great constellation Argo, it seems to 
command, while standing slightly aloof from, the dazzling 
array of all stellar ranks spanning the heavens from the 
Greater Dog to the Cross. But since its bluish-white rays are 
almost undimmed by absorption, the probability is strong that 
their intensity largely exceeds the solar proportion of luminosity 
to mass. 

Now Dr. Elkin's failure to detect any parallactic shifting 
in Canopus obliges us to suppose it at such a distance that 
its light needs at least sixty-five years to reach us ; how much 
longer, it is impossible to tell. At this minimum remoteness, 
our sun would shrink to a 7'5 magnitude star ; it would be 
one of the dense shoal of telescopic objects imperceptible to 
unaided sense, and individualised only by the industry of 
astronomers. But 2,500 stars of 7*5 magnitude give only 
the light of one Canopus, whence it follows that Canopus is 
certainly brighter, and may be very greatly brighter, than 
2,500 Sims hke ours. There is only one way of escaping 
from this startling conclusion. It is possible, though far from 
probable, that both Dr. Elkin's reference-stars are physically 
connected with the brilliant object they lie near ; in which case, 
his result was of course null and void. This doubt, however, 
will shortly be set at rest through Dr. Gill's remeasurement 
of Canopus with a fresh and much wider pair of stars. 

The ninth star tested at the Cape in 1881-8 was fi Cen- 
tauri, one of the so-called southern Pointers, and taking rank 
among the lower grades of the first magnitude. Dr. Gill's 

THE DISTAimE£^l^g3aS><TAES 311 

negative result made it all but certain that the considerable 
absolute parallaxes previously derived from meridian observa- 
tions of the star were illusory ; but here again, a new series 
of measures by the same eminent investigator must prove 

The time had now come when a change in the system upon 
which inquiries of this kind were prosecuted seemed feasible. 
Hitherto, observers had been content to select the most pro- 
mising subjects for their experiments, without any regard to 
the co-ordination of results. The outcome was a collection of 
detached statements as to stellar distances, interesting, each 
by itself, in a high degree, yet incapable of being combined 
for the purpose of any general conclusions. So long ago as 
1853, Dr. Peters had pointed out that what was needed for 
obtaining a fundamental acquaintance with the structure of 
the sidereal world was not so much the determination of 
exceptional parallaxes, as the steady compilation of data for 
some well-grounded inference relative to the distances of 
defined star-classes,^ But it was not until thirty years later 
that it became possible to act on the suggestion. 

Encouraged by the success of the work just accomplished, 
Dr. Gill proposed, January 11, 1884, a scheme for dealing 
with the problem of star-distances in its widest bearings. 
Two ' great cosmical questions,* he saw, pressed for answers, 
which might be obtained by the judicious distribution of some 
years' continuous labour. They are : 

1. * What are the average parallaxes of stars of the first, 
second, third, and fourth magnitudes respectively, compared 
with those of lesser magnitude ? ' 

2. ' What connection subsists between the parallax of a 
star and the amount and direction of its proper motion, or 
can it be proved that there is no such connection or relation ? ' "^ 

Some advance has already been made towards procuring 
at least partial replies. A plan was concerted by which 
Dr. Elkin, now in charge of the new Yale College heliometer, 
undertook the measurement of a considerable number of 

* M&moire$ de St. Pitershourgy t. vii. p. 149. 
- Mtmoir$ B, Astr. Soc. vol. xlviii. p. 191. 


representative northern stars, whDe Dr. Gill dealt with a 
corresponding southern list at the Cape, where a seven-inch 
heliometer, constructed for the purpose by Messrs. Repsold 
with the utmost perfection of instrumental art, was erected in 
the summer of 1887. In the previous determinations, the 
Dunecht heliometer, become, by purchase from Lord Crawford, 
th^ private property of Dr. Gill, had been employed ; but its 
aperture of only four inches, restricting the choice of com- 
parison-stars to brighter objects than were always to be found 
in the most advantageous situations, made it unsuitable for 
carrying out the comprehensive plan later decided upon. 

It included an investigation of the parallaxes of all first- 
magnitude stars — ten in the northern, ten in the southern 
hemisphere. This done, the average distance of stars of the 
highest Ught-rank becomes known, no longer by inference or 
guessing, but by direct measurement. A scale-unit for the 
stellar universe will then, at last, be available. For, once we 
know the distance in billions of miles or light years corre- 
sponding to the first magnitude — the distance, that is, at 
which a * mean star * would shine with about the lustre of 
Spica or Capella — the distances corresponding severally to the 
lower magnitudes follow as a matter of course. They are 
linked together by an invariable proportion. We have 
already explained what is meant by the * hght-ratio,' * but it 
may here be repeated that a star of any given magnitude is, 
by definition, one 2*512 times brighter than a star of the 
magnitude next below, and 2*512 times less bright than a 
star of the magnitude next above it. But, since hght varies 
inversely as the square of the distance, any star removed to 
v^ 2*512 =1'585 times its actual distance, would show exactly 
one magnitude fainter than it did before. This number, then, 
1*585, the square-root of the hght-ratio, may be designated 
the * distance-ratio.' It represents the difference of distance 
corresponding to a difference in hght of one stellar magnitude. 
The relative mean distances of the various classes of stars 
are then known; to render them absolute, we only need to 
ascertain the real mean distance of any one of those classes. 

» See ante, p. 20. 


It is true that, within each class, enormous disparities 
exist. Small stars, comparatively near the earth, take their 
stand on the same level of apparent brightness with indefinitely 
large, but indefinitely remote bodies. What is invariable for 
each magnitude is the proportion between real brilliancy and 

the square of the distance. Symbolically expressed, ^^ is 

constant. That is to say, photometric uniformity results 
from a certain balance being struck between remoteness and 
light-power, by which the effect of equality is produced. The 
law, however, connecting average distance with apparent 
lustre is not invalidated even by the limitless variety included 
in the above expression. The extremes are vastly wide apart, 
but the mean remains, for all practical purposes, the same. 
It should, nevertheless, be always clearly borne in mind that 
the conclusions thus obtained are general, and should only 
be generally applied. Eeferred to particular cases, they are 
utterly fallacious, and more likely than not to mislead. 

The interest attaching to Dr. Elkin's deduction, in 1888, 
of a mean parallax of 0''-089 for the first-magnitude stars of 
the northern hemisphere,^ can now be appreciated. Even as 
half a result, it is worth attentive consideration. 

It establishes provisionally, at a distance of 36 J light-years, 
the first halting- place for explorations of sidereal space. Thus 
inconceivably remote, on an average, are the brightest and, 
we must add, the nearest stars. Our sun, thus placed, would 
sink below ordinary naked-eye visibility ; it would be of 6*6 
magnitude. So that the conclusion of its insignificance among 
its fellow suns is rendered at least plausible. 

On the scale measured at Yale College, the mean distance 
of stars of the second magnitude is 58 light-years (parallax 
0''-056) ; stars of the third magnitude are at 92 light- 
years (parallax 0"'035), and so on; the invariable ratio of 
1*585 regulating the increase of distance and decrease of 
magnitude for each descent of one step (provided only that 

* Sid, Mess. Nov. 1888, p. 395. The resalt actaaUy obtained was 0"-085, 
relative to the compariBon-starSf for which the. observations indicated a prob- 
able average parallax of 0'''04. This amount, added to 0"'085, gave 0"'089 as 
the absolute mean parallax of the ten stars measured. 


light suffers no diminution of intensity in interstellar space). 
When we get down to the sixteenth magnitude, which is about 
the minimum visibile in the largest telescopes (the Lick re- 
fractor may perhaps go a magnitude lower still), we find the 
theoretical light-interval lengthened to 36,000 years; but 
there is no certainty that any such far- travelled rays do, in 
point of fact, reach us. The regular progression of distances 
may not extend so far. It must stop somewhere, if the 
stellar system be —as we have reason to think it is — of finite 
dimensions ; at what particular magnitude, however, the 
break occurs, it would at present be futile to conjecture. All 
that can be said is that, distance becoming at length elimi- 
nated as a factor of magnitude, the differences of the faintest 
stars represent, chiefly or solely, real inequahties in shining. 
There may possibly, for instance, be no ' mean distance ' cor- 
responding to the sixteenth magnitude. The stars of that 
rank would not then, on the whole, be further off than those 
of the rank next above them, but would, on the whole, possess 
only ^ of their real light. This must be the case — so far as 
we can see — at some stage of the descent into the abysses 
around us. 

Of the ten stars measured at Yale, Procyon, with a parallax 
of 0"-266, giving a light-journey of 12-3 years, was found to be 
the nearest to the earth. Altair, at a distance of 16-4 light- 
years (parallax 0"-199), came next ; Aldebaran, at 28 light- 
years (parallax 0"'116), third; Regulus, with a parallax of 
0''-093, might be termed a mean first-magnitude star at mean 
distance. Of the remainder, four gave no reliable indications 
of possessing any sensible parallax. That is to say, the 
concluded value — which was negative for Betelgeux and 
Deneb, positive for Arcturus and Vega — was in each case quali- 
fied by a probable error larger than itself. As to Vega, there 
is still some doubt, a considerable parallax relative to its well- 
known optical attendant having been found for it by several 
previous observers; but that a Orionis, a Cygni, and Arc- 
turus, are plunged in depths of space unfathomable by any 
method yet brought into use, may be admitted without hesi- 
tation. The excessive remoteness of Arcturus, especially, 


enables us to recognise in it perhaps the most stupendous sun 
within our imperfect cognisance. For, although less lustrous 
than Canopus, it is probably (judging from the character of 
its spectrum) much more massive proportionately to its 
light, while the rate of its movement can only be termed 
portentous. Assuming the accuracy of Dr. Elkin's nominal 
parallax of 0''*018, its velocity across the line of sight alone 
must reach 872 miles a second, 380 being the utmost that 
our sun can generate in a body reaching its surface from 
infinite space. 

So closely and so consequentially have advances in this 
arduous branch followed the growi^ of improvement in 
heliometers, that direct visual measurements for the purpose 
with any other instrument might almost seem waste of labour. 
But direct visual measurements are no longer the only prac- 
ticable ones. By the energy of Professor Pritchard, the 
determination of stellar parallaxes by photography has been 
converted into a working reaUty of incalculable value and 
promise. His first experiment was with the classic 61 Cygni, 
of which 330 separate impressions, obtained in 1886-7, fur- 
nished the materials for 30,000 measurements, or ' bisections ' 
of star-images.^ For the immediate purpose, these extra- 
ordinary pains were largely superfluous; but they had the 
ulterior object, fully attained by their means, of estabUshing 
the credit of a novel and unfamiliar method. Not only is the 
resulting parallax of 0"'43 for 61 Cygni of the highest au- 
thority, but the most delicate of all astronomical inquiries 
was at once, with the full assent of experienced judges, ad- 
mitted to be within the full competence of the photographic 

Similar determinations have since been made of Polaris, 
a, )9, and fi Cassiopeiae. The results will be found in the 
Table of Parallaxes appended to this volume. At the distance 
of Polaris, our sun would be discoverable with an opera-glass 
as a 7*5 magnitude star ; in the place of fi Cassiopeise it 
would sink to 8*4 magnitude. The pole-star hence emits the 
light of 158, fi CassiopeiflB of nineteen suns, while the large 
' MontJUy Notices^ vol. xlvii. p. 87. 


proper motion of the latter comes out, in linear velocity, 
363 miles per second ! 

The advantages of photography for stellar parallax- work are 
manifold. Perhaps the chief of them is the nearly indefinite 
power of control afforded by it. Any star on the plates, 
situated at all near the prolongation of the major axis of the 
parallactic ellipse (in other words, with a tolerably large 
* parallax-factor '), may be used as a point of reference. Com- 
parisons can thus be multiplied almost at pleasure, and inferred 
displacements with regard to one star checked by recourse to 
another, duplicate plates being at hand for additional safety. 
By the proper use of such safeguards, delusive results can be 
all but certainly excluded. Moreover, relative parallax becomes 
virtually absolute when comparisons are made with a great 
number of stars, most of which are presumably too remote to 
complicate the result by perspective movements of their own. 

The peculiar province of photography is with stars too 
faint to be conveniently dealt with by visual means : for the 
images of the brightest ones, over-exposed through the 
necessity of giving the small stars in their neighbourhood 
time to imprint themselves, become diffused into blurred discs 
beyond the power of accurate bisection. Thus, the investiga- 
tion of stellar parallax relatively to magnitude falls naturally 
to workers with the heliometer, whDe its relation to proper 
motion can best be elucidated with the aid of the photographic 

The main object just now of inquirers in this branch is 
to obtain a wider basis for general conclusions regarding the 
distances of the stars. For this purpose it is more important 
to secure a considerable number of parallaxes reasonably well 
determined than a few reduced by scrupulous care within the 
narrowest possible bounds of error. Research in this sense 
is already well on its way. It has been pointed out by 
Mr. Monck ^ that the average parallax of stars of the second 
will prove a more reliable datum than that of stars of the 
first magnitude, both because they are more numerous, and 
because they are more nearly equal in brilliancy ; and there 

» Sid, Mess, Feb. 1889, p. 62. 


is a prospect that this datum will before long be provided, 
at least in an approximate form, by the Oxford plates rein- 
forced by Cape measures. Its actual ascertainment will con- 
stitute a step of the very highest importance. 

For the present, we have only to ask ourselves, what are 
the indications derivable from the work so far as it has gone ? 
They are decidedly, in the first place, against the existence of 
any large parallaxes. It is, of course, still amply possible that 
stars may be found much nearer to the solar system than 
a Centauri, but their discovery is growing every year less and 
less probable. Sir Robert Ball ' examined some years ago 
about 450 objects in a manner which, though summary, would 
have sufficed to bring to view any parallax of a single second 
of arc. None was forthcoming. His list included a number 
of red and variable stars. Nova Cygni, Webb's planetary 
nebula, and the Wolf-Rayet gaseous stars in Cygnus ; and it 
may be noted in passing that no star with a banded, or a 
bright-Kne spectrum has yet exhibited the slightest tendency 
towards parallactic displacement. Until, however, many more 
efforts have been made for its detection, much significance 
cannot be attached to its absence. The best chance of success 
would perhaps be with Mira Ceti, the well-determined proper 
motion of which is suggestive of proximity. 

The cardinal truth emerging from these inquiries is that 
of the extreme isolation of the solar system. A skiff in the 
midst of a vast, otherwise unfurrowed ocean, is not more 
utterly alone. About the same proportion would be borne by 
an oasis one mile across to a desert twenty times as extensive as 
the Sahara, that our sun with his entire planetary household 
bears to the encompassing void of space. The enormity of its 
blank extent is strikingly illustrated by Father Secchi's remark 
that the period of a comet reaching at aphelion the middle 
point between our sun and the nearest fixed star, would be of 
one hundred million years ;^ and by recent measures, the 
nearest fixed star has been pushed further back into space by 
one quarter the distance assigned to it when he wrote. Yet 
the sun is no isolated body. To each individual of the un- 

* Dunsink Observations^ vol. v. « Les Etoiles, t. ii. p. 146. 


numbered stars strewing the firmament, down to the faintest 
speck of light just shimmering in the field of the Lick refractor, 
it stands in some kind of relationship. Together, they master 
its destiny, and control its movements. Independent only so 
far as its domestic affairs are concerned, it is bound, as a star 
to the other stars, by influences reaching efficaciously across 
the unimaginable void which separates it from them. The 
outcome of those influences in the translatory motion of the 
solar system, we shall consider in the next chapter. 




The study of the stars inevitably leads us to consider the ad- 
vancing movement in the midst of theni of the sun and its at- 
tendant train of planets. There can be no reasonable doubt — 
and the thought is an astounding one — that we are engaged 
on a voyage through space, without starting-point or goal that 
we can know of, but which may prove not wholly uneventful. 
Its progress may possibly bring about, as millenniums go 
by, changes powerfully influential upon human destinies; 
nay, an incident in its course may, at any time, by the in- 
scrutable decree of Providence, terminate the terrestrial ex- 
istence of our race, and consign the records of its civilisation, 
in dust and cinders, to the arid bosom of a dead planet. A 
curious sense of helplessness, tempered, however, by a higher 
trust, is produced as we thus vividly reaUse, perhaps for the 
first time, how completely we are at the mercy of unknown 
forces — how irresistibly our little ' lodge in the vast wilderness ' 
of the universe is swept onward over an annual stretch of 
perhaps five hundred millions of miles, under the mysterious 
sway of bodies reduced by their almost infinite distances to 
evanescent dimensions. 

But, as things are constituted, the translation of the sun's 
household is a necessity, albeit one of startling import to our- 
selves. The stellar system is maintained by the balance of 
forces, and motion is the correlative of force. As a star among 
stars, the sun can only maintain a separate existence by con- 
tributing its share to those harmonies of movement by which 
* the heavens show forth the glory of God.' Destruction would 
be the eventual penalty of even a moment's immobility. A 
penalty, indeed, which might not be exacted until after the 


lapse of many millions of years. It may reasonably be 
assumed that a Gentauri exercises upon the sun the strongest 
attraction of any individual star ; but a collision would ensue 
very tardily upon abandonment to its influence. The sun (if 
undisturbed by competing pulls) would fall from a position of 
rest towards its next neighbour, less than the third of an inch 
in one month ; the second month would see despatched nearly 
a full inch of the journey of twenty-five billions of miles ; and 
although the acceleration would of course grow more rapid as 
the distance diminished, close upon fifteen million years should 
elapse before the fires of sun and star, probably become extinct 
during their gradual approach, could be rekindled by the 
catastrophe of their impact. 

There is then an a priori certainty that the sun moves ; 
and assurance on the point is rendered doubly sure by 
inferences from observed facts. For besides their annual 
parallax due to the earth's motion round the sun, the stars 
have a * secular ' or ' systematic ' parallax depending upon and 
attesting the reality of the sun's motion round an unknown 
centre. Let us see how this systematic parallax can be 

If the sun alone were in motion, and the stars at rest, the 
results in perspective displacements would be simple and un- 
mistakable. Each star would appear to travel backward along 
a great circle of the sphere, passing through the two points to- 
wards and from which the sun's course was directed. So that 
there would be the semblance of a general retreat from the 
* apex,' or solar point de mire, coupled with a thronging-in from 
all sides towards the opposite point, or ' anti-apex.' For each 
particular star, the amount of displacement should vary, in- 
versely as its distance from ourselves in space, directly as the 
sine of its angular distance from the apex. Hence, if the 
annual parallax of even one such sensibly shifting star were 
determined, not only the rate, in miles per second, of the 
solar progression would at once follow, but the parallax of 
every other sensibly shifting star in the heavens could be de- 
duced by a simple calculation from the relative quantity of its 
apparent movement. 


But the stars are not at rest. They have movements of 
their own, greatly swifter, in many cases, than that of the sun. 
Perspective effects are thus to a great extent masked. Yet 
they subsist. It is mathematically certain that every star, 
whatever its own course or speed, reflects the sun's motion in 
the strict measure of its position with regard to it. What are 
called the * proper motions ' of the stars are then made up of two 
parts, one real, the other apparent. They include a common 
element, the separation of which from the heterogeneous ad- 
mixtures disguising it, constitutes the problem to be solved. 

With the instinctive appreciation of genius, Herschel went 
straight to the heart of the matter. What had to be done, he 
saw clearly, was to find out the direction which should be given 
to the sun's course, in order to make it account for as large a 
proportion as possible of the sum-total of stellar movements. 

* Our aim must be,' he wrote in 1805, * to reduce the proper 
motions of the stars to their lowest quantities.' * And again : 

* The apex of the solar motion ought to be so fixed as to be 
equally favourable to every star.' But how is this to be done ? 
Very simply, if we only consider, as Herschel did, a few of the 
brightest stars. 

Take, for example, four stars with conspicuous movements, 
two in the northern, two in the southern hemisphere — namely, 
Vega, Capella, Sirius, and FomaUiaut. The great circles, of 
which each yearly describes a minute arc, traced backward on 
the sphere, very nearly intersect in a single point situated in 
the constellation Hercules.^ Had we only the motions of those 
four stars to consider, we should accordingly infer without 
hesitation the * sun's way ' to lie thitherward. Nor should we 
be very far wrong. The most refined modern determination 
of the solar apex, founded upon the motions of several thousand 
stars, differs by only five degrees from the result of the ex- 
tremely summary proceeding just indicated. 

The graphical method, however, is evidently applicable 
only to a very restricted stock of data. When a crowd of stars 

» PML Trans, vol. xcv. p. 248. 

* This was remarked by Klinkerfups, GUttingisclie Nachrichtent 1873, 
p. 350. 



have to be taken into account, the points of intersection of 
their respective circles of motion become spread over too wide 
an area for a * mean apex ' to be struck out fairly between 
them, even by the exercise of a judgment as discriminating as 
that which in 1783 led Herschel to set the present goal of 
solar travel in the vicinity of X HercuUs. The accumulated 
facts must then be dealt with by a method at once stricter 
and more comprehensive. A glance at the nature of the task 
in hand easily suggests to a mathematician what that method 
should be. 

The proper motions of the stars give, as already hinted, 
the plainest evidence of individuality. The lines traced by 
them on the sky run in all possible directions. But a sub- 
stratum of regularity underlies this seeming confusion. A 
mere inspection of the signs plus and minus, signifying respec- 
tively east and west, and north and south, attached in cata- 
logues to the components in right ascension and declination of 
stellar movement, suflSces to show a general prevalence of law 
through the unequivocal tendency of the signs to vary together 
in passing from any one to an adjacent region of the heavens.* 
At a coup d'(xily Argelander fixed the point from which this 
under-current of motion flowed, and so gave an improved 
apex for the course of the sun confirmed in general by all 
subsequent research.^ It is then clear, in the first place, that 
no movement possibly assignable to the sun can explain all 
stellar displacements ; a large residuum being real, and there- 
fore by no ingenuity to be got rid of. While in the second 
place, the nearer the truth is approached as regards the direc- 
tion and amount of the sun's motion, the smaller obviously 
this residuum will be. In other words, the most probable 
value of the solar motion will be that which renders the * sum 
of the squares of the residuals ' of stellar motion a minimum. 
But why the sum of the squares, and not the simple arith- 
metical sum of the outstanding proper movements ? It needs 
only common sense, aided by the most elementary geometry, 
to get a sufficient insight into the reason. Any one can see, 

* Stone, Monthly Notices, vol. xxvii. p. 239. 

* M^moircs pr^se7iUs a VAcad. St. P6tersbourg, t. iii. p. 5G9. 


tvith the help of a pencil and a piece of paper, that if a line 
be divided into two segments, and squares be constructed on 
the segments, the sum of those squares will be the least pos- 
sible when the line is equally divided, and will increase con- 
tinually with the inequahty of the segments. This simple 
fact gives the clue to the principle of * least squares.' Its 
object is to elicit such a quantity as will make the outstanding 
errors of observation, or any other kind of residuals, as small 
as possible all round. Not merely small taken in the aggre- 
gate, but reduced impartially to a uniform level of insignifi- 
cance- Under these circumstances, as we have seen from the 
consideration of our divided line, the sum of their squares 
will be a minimum ; and it can be mathematically demon- 
strated that the most probable result in such investigations as 
are susceptible of this kind of treatment is arrived at when 
the condition of ' least squares ' is fulfilled. 

This mode of attack upon the problem of the sun's trans- 
lation was first employed by Argelander in 1837. Assuming 
l^rovisionally the correctness of Herschers apex, he proceeded 
to compute for each of 390 stars with ascertained proper motions 
the lines along which those motions should proceed if due 
to systematic parallax alone. Their deviations from the pre- 
scribed directions gave him * angles of error,' which, placed in 
the category of casual errors of observation, and treated 
by the method of least squares, indicated a corrected apex, 
such that by its adoption the sum of the squares of the 
differences between what was calculated and what was ob- 
served — that is, between the purely parallactic drift of the 
stars and their actual displacements — was reduced to the least 
possible amount. The solar movement was, in a word, so 
fixed as, in Herschel's phrase, * to be equally favourable to 
every star.' According to this determination, the sun's way 
lies towards a point in right ascension 260° 51', north decli- 
nation 31° n\ occupied almost exactly by a sixth- magnitude 
star numbered in Piazzi's catalogue 143 in the XVIP** hour.^ 
^or is there any sign that Argelander's confidence in its sub- 
stantial accuracy was misplaced. 

• M^inoircs pr^senh^s^ t. iii. p. 590. 

Y 2 


An important modification of his method was introduced 
by Sir George Airy in 1859.^ Abolishing the conception of a 
spherical surface of reference, he defined the linear move- 
ments in space of the sun and stars with regard to three 
directions at right angles to each other (* rectangular co- 
ordinates '). No assumption of any kind was then needed ; 
the subject was treated with the utmost strictness and gener- 
ality, and some possible causes of error were removed. Airy's 
had many points of theoretical superiority over Argelander's 
method. That, however, of introducing the consideration of 
the quantity of each star's movement was to a great extent 
counterbalanced by the necessity which it involved of adopt- 
ing precarious suppositions as to the distances of the classes 
of stars employed. The apex for the solar movement resulting 
from the consideration of 113 stars was situated in E.A. 
261° 29', Dec. + 24° 44' ; while Mr. Mam's similar treatment 
of 1165 stars shifted it to R.A. 263° 44', Dec.-f 25°.» 

In the latest, and most thorough investigation of this great 
subject, by M. Ludwig Struve,* Airy's method was adopted. 
The incitement to undertake a task rendered formidable by 
the very wealth of the materials at his disposal was afforded 
by Auwers's fresh reduction of Bradley's Greenwich observa- 
tions. From a comparison of the star-places thus authorita- 
tively determined for 1765, with those given in the St. Peters- 
burg catalogue for 1855, a list of 2,814 proper motions was 
derived, of which 2,509 were available for M. Struve's purpose. 
Among the stars for various reasons excluded were the seven 
swiftest travellers, as unduly affecting the result through 
motions no doubt mainly original. 

As the outcome of this exhaustive discussion, the apex of 
the solar motion was placed in B.A. 273"* 21', north declina- 
tion 27° 19', and a rate was assigned to it such that the space 
traversed in a century, viewed square from the average dis- 
tance of a sixth-magnitude star, would subtend an angle of 
4"-86. Admitting that stellar distance varies inversely as the 
square root of stellar brightness, hence that stars of the first 

* Memoirs R. Astr. Soc. vol. xxviii. p. 143. * Ibid. vol. xxxii. p. 27. 

' Mdmoires de St. Piftertbourg, t. xxxv. No. 3, 1837. 


are, on an average, only one-tenth as remote as stars of the 
sixth magnitude, we can, with the help of Dr. Elkin*s mean 
parallax for the former class, translate this angular into linear 
velocity. It comes out 14^ miles a second. 

Well-nigh the whole of the stars visible to the naked eye 
in the northern hemisphere concurred in M. Ludwig Struve's 
determination. No principle of selection was employed ; they 
were taken as they came ; exclusion was only resorted to in 
the few cases where misleading peculiarities were obvious. 
But the surest proof of success was derived from the applica- 
tion of a test which had served only to impair confidence in 
previous results — the test, namely, of residual errors. The 
transport of the solar system must, unless it be shared by the 
observed stars, add, in the proportion of its velocity, to their 
apparent movements ; the subtraction from them of the 
common drift it produces and is disclosed by, should then 
leave those apparent movements materially diminished. But 
this had hitherto been the case only to a very unsatisfactory 
extent. The withdrawal, however, of the parallactic element 
discriminated from them by M. Struve, at once reduced the 
sum of the squares of the * corrected ' proper motions nearly 
to one-half its original amount, thereby testifying emphatically 
to its own genuineness. 

Combining his with fifteen prior results, M. Struve finally 
concluded for an apex in E.A. 267°, declination -f 81°. This 
position, as he said, cannot be far wrong. Apart from a 
source of uncertainty to be presently adverted to, the aim of 
the sun's motion may now be regarded as approximately known. 
Above a score of investigators have agreed in directing it 
towards a restricted area in the heavens marked by the 
outstretched left arm of the asterismal Hercules. Varying 
methods have been employed ; hypotheses of sundry kinds 
have been adopted; diflferent stars have been appealed to, 
with the same general result. The conviction of its truth 
becomes irresistible when we find that the entire stellar 
heavens pronounce in its favour. Indications of an identical 
slow steady flow of motion aicay from the constellation 
Hercules, and towards the constellation Columba, come from 


the southern, as well as from the northern hemisphere. 
Galloway's experiment in 1847, giving an apex in R.A. 260"", 
Dec. -f 34° 23',' was striking, but could hardly be regarded as 
decisive, owing to the doubtful character of the data available 
to him. But M. de Ball's research thirty years later,* while 
of similar upshot, was not open to the same reproach, a small, 
though sufficient stock of authentically determined southern 
proper motions having in the meantime accumulated. And 
his conclusion was virtually re-affirmed by Mr. Plummer, with 
the help of 274 stars from Mr. Stone's ' Cape Catalogue for 
1880.' » 

But when, from the direction^ we attempt to pass to the 
amount of solar motion, the case becomes widely different. 
Flagrant contradictions abound. Estimates of velocity range 
at large between five and 150 miles a second ; the criteria of 
truth are at the mercy of individual judgment. The cause of 
these discrepancies lies in the uncertainty still prevailing as to 
the distances of the stars. Quite irrespectively of the remote- 
ness of the objects whose perspective displacements serve as 
the index to it, the line of our advance through space can be 
searched out ; but its rate can only become known when we 
know how far ofif the displaced objects are. The relative 
distances, however, assumed by most observers in this field 
being for the most part untrustworthy, the resulting angular 
velocities of the sun are no less so. And angular can of 
course be converted into linear velocity only through acquain- 
tance with absolute distances. 

It is nevertheless tolerably certain that the solar pace has 
nothing headlong about it. We are not whirled in the train 
of such a stellar projectile as 1830 Groombridge or f Toucani. 
Our condition, were it so, would be betrayed by unmistakable 
tokens. Everything, on the contrary, suggests the inference 
that our sun is among the more sedately moving stars. The 
ascription to it of a very high velocity entirely unwarranted by 
facts has had its spring in erroneous assumptions. From a 

' Phil, Trans, vol. xxxvii. p. 79. 

« \\ ochemchrift fUr Astr, lid. xx. p. IfiO. 

■ Memoirs li. Astr. Soc. vol. xlvi p. 341. 


statistical survey of the elements of motion in 1167 stars, Mr. 
Stone concluded their real to exceed their parallactic move- 
ments in the proportion of four to three ; * and there is much 
reason to believe this estimate tolerably correct. A combina- 
tion with M. L. Struve's of three previous values placed the 
linear velocity of the solar system at about sixteen miles a 
second. On the plan of inquiry heretofore considered, it is 
scarcely possible, for the present, to get nearer to truth on the 
point ; but another is rapidly becoming available. 

We have elsewhere explained the principle of spectroscopic 
determinations of motion.* Their peculiar value consists in 
their independence alike of distance and of visible displace- 
ment. Referring to movements visually imperceptible, they 
complete knowledge of stellar velocities by giving their other- 
wise unknown 'radial components.' Apart from this won- 
derful application of the spectroscope, the real lines of travelling 
of the stars could never have been ascertained, since we can 
immediately discern only that part of their motion lying across 
the line of sight, which, in individual cases, may be all or 
none. By the spectroscopic revelation, however, of motion iti 
the line of sight, the missing element is supplied, precise and 
particular knowledge may be had for the asking, and the stars 
pursue their journeys under astronomical scrutiny, no longer 
as mere flitting bright specks on the surface of an imaginary 
sphere, but as suns in space, each with its own secret in 
reserve, and each contributing to swell and deepen the 
marvellous harmony of the whole. 

The effects of recession and approach on the light emitted 
by moving objects being physical and real, they remain 
unimpaired by distance. Out at the verge of the sidereal 
system, or close at hand within our own atmosphere, they are 
the same for the same velocities, and can, with a sufficient 
light-supply, be detected with equal facility. Hence their 
special applicability to the problem of the sun's speed. To 
determine it with very approximate accuracy, it will only be 
necessary to compare the average radial celerity of a good 
number of stars lying in front of the sun's way with that of 

' Monthly Notices, vol. xxvii. p. 239. * Hist, of Astr. pp. 245, 440. 


others he is leaving behind. Movements of approach must, 
on the whole, predominate in the one direction, movements of 
recession in the opposite, half the mean difference, elicited by 
appropriate processes of computation, representing the rate of 
transport of our system. The method, however, cannot be 
successfully employed with such data as were, until lately, the 
only ones to be had. In a matter of such excessive delicacy as 
the measurement of minute shiftings of Unes in stellar spectra, 
eye-observations are subject to too many and too grave dis- 
turbances for implicit reliance to be placed on them. They 
have done their work in showing the validity of the principle 
and giving preliminary indications of no small importance ; 
and now the turn of the photographic camera has come. The 
quality of the materials furnished by it under Professor Vogel's 
able treatment leaves Uttle or nothing to be desired ; when 
their quantity has been sufficiently augmented, the rate, in 
miles per second, of our transport through space will become 
known with ease and certainty. 

Some experimental attempts have indeed already been 
made to extract the desired information from visually 
determined line-of-sight movements ; and M. Homann's 
especially, notwithstanding the defective nature of the infor- 
mation at his command, may serve as an index to the future 
value of the method. Forty-nine of the brightest stars, 
spectroscopically determined between 1878 and 1884, gave 
him a goal for the sun's way, no longer in the accustomed 
situation in Hercules, but transferred some four hours east- 
ward to very near the place of 61 Cygni.* This new apex, it 
may be remarked in passing, lies in one of the densest parts 
of the Milky Way. The corresponding velocity of fifteen miles 
per second agrees so closely with M. Ludwig Struve's estimate 
as to encourage the persuasion that it needs little amend- 

Thus, both the direction and rate of transport of our system 
are not only included in the category of things knowahle, but 
there is every prospect of their becoming known with more 

^ Homann, Astr. Nach, No. 2714 ; Schonfeld, V. J. S. Astr. Ges. Jahrg. xxi. 
p. 58. 


and more satisfactory exactness in the immediate future. All 
that is needed is a closer and a wider application of means 
already in the hands of astronomers. Still our curiosity will 
not even then be satisfied. The value of the two items of 
information within our reach is indeed incalculable. They are 
a sine qua nan for the furtherance of inquiries into stellar 
mechanics ; are they to be a ne plus ultra as well ? 

The sun, we are well assured, is not travelling along a 
straight line. The universality of gravitation makes rectilinear 
movement next to impossible, since no cosmical body can traverse 
space under the sole guidance of its own primitive velocity. It 
is true that, supposing primitive velocities altogether abolished 
(and we know of no reason why they should necessarily exist), 
any number of bodies might be united into a system endowed 
only with pendulum-like motions. The sun and stars might 
thus, by an abstract possibility, be totally devoid of advancing 
or circulatory movements, each swinging for ever to and fro 
through their common centre of gravity. But it is practically 
certain that this plan is not realised in the sidereal system. 

The path of the sun is then a curve, but a curve most 
likely of such vast proportions as to remain for ages indis- 
tinguishable from a right line. Strictly speaking, however, 
its direction is continually changing ; the apex of to-day will 
not be the apex of to-morrow ; still less will it be the apex 
of a million years hence. Yet in a million years it may quite 
conceivably not have shifted from its present place in the sky 
by more than the width of the full moon ; and our best deter- 
minations still fall far short of the accuracy which would enable 
us to detect a change of half a dozen times that amount. 
Directly, that is to say ; indirectly, a much more insignificant 
alteration might disclose itself. It can easily be explained 

Pond, who succeeded Maskelyne as Astronomer Eoyal in 
1811, made the remark that the sun's motion must produce a 
kind of secular aberration of light, by which the stars are 
permanently displaced from their true positions.* The effect 

» Liagre, Bull, de VAcad. Bruxelles, t. viii. p. 168, 1859 ; 0. Struve, 
Mimoires, St. Pfitersbourg, t. v. p. 106, 6* S6rie. 


of ordinary, or what we may call annual aberration, is to make 
them appear to describe little ellipses, the semi-axes of which 
depend upon the ratio of the velocity of light to the velocity 
of the earth in its orbit. But the sun*s orbital movement 
being conducted, so far as experience yet goes, in one direction, 
the aberration due to it is in one direction too, and is hence 
constant, and for the present beyond the reach of observation. 
It is, however, constant only so long as the movement pro- 
ducing it remains sensibly so. As the latter changes, it will 
change too, and may in this way be brought within the domain 
of human cognisance. For upon the acceleration, retardation, . 
or deflection of the sun's movement systematic displacements 
among the stars should ensue, the nature of which would at 
once betray their origin. 

The total amount of this secular aberration may be 
roughly stated as one second of arc for every mile per second 
of the sun's velocity. Hence, stars 90^ from the solar apex 
are pushed forward towards it by perhaps 15", the eflfect upon 
other stars diminishing with the sines of their distances on 
the sphere from the same point. These aberrational can be 
distinguished from the parallactic displacements flowing from 
the same source, by their indifference to remoteness in space. 
Stars far and near, bright and faint, swift-moving and tardy, 
are equally affected by them. But while it is quite certain 
that visual disturbances of this kind take place, and to an 
extent possibly greatly in excess of that above assigued to 
them, their interest must for a long time remain purely 
theoretical. Indeed, it may well be that the modifications 
rendering them sensible and instructive will proceed with 
such exorbitant slowness that not even astronomical patience 
will avail to unmask them. 

We do not know the plane of the sun's orbit — only the 
direction of one line in it. And that line, pointing towards 
the constellation Hercules, makes an angle of about 60° with 
the sun's equator. Thus, the solar movements of rotation 
and translation would seem to be unrelated one to the other ; 
and the same remark applies to the planetary revolutions 
conducted, on the whole, along levels of space differing very 



The movement 

little from that of the greater globe's axial movement. Our 
whole system is then driven obUquely upward by a power 
which, taking no apparent account of its domestic economy, 
owned doubtless an origin totally disconnected from that of 
gyrations given, through its influence, the helicoidal shape 
illustrated in fig. 48. 

The sun's course, being inclined some twenty-eight degi-ees 
to the central plane of the Milky Way, is at present gradually 
removing it from that stupendous collection. This, however, 
does not necessarDy imply real separateness. 
of withdrawal actually progress- 
ing may be only temporary, in 
the sense that, after countless 
millions of years, it will be com- 
pensated by a return movement 
of approach. It is difficult to 
conceive that the> combined at- 
tractions of the galactic myriads 
can, in the long run, be resisted. 
The most probable supposition 
as to the situation of the centre 
of force swaying our system, is 
that it lies somewhere in the 
cloudy zone which so enhances 
the mysterious beauty of our 
skies. If the orbit we are pur- 
suing be approximately circular, 
then its centre must be divided by a quadrant of the sphere 
from the apex — it must lie somewhere on a great circle 
of which the apex and anti-apex are the poles. Now this 
great circle cuts the Milky Way at two opposite points in 
Perseus and Hydra, and there, accordingly, two alternative 
centres of the solar motion might be looked for. Argelander 
chose as the more promising position the spot marked by 
the great cluster in the sword handle of Perseus ; ^ but 
the conjecture made no pretension to scientific authority, 
and the postulate upon which it was based, of the sun's 

' M^nwires pr^sentis, St. P^tersbourg, t. iii. p. >02. 


48.— The Earth's Motion in 


path being at all nearly circular, is in truth of a highly 
precarious nature. 

We are even ignorant whether the translation of our 
system towards the constellation Hercules represents a primary 
or a secondary order of stellar revolution. It perhaps merely 
indicates the interstitial movement appertaining to the sun as 
a member of a restricted group of stars, the common transport 
of which proceeds undetected in a totally dififerent direction. 
Hence the possibility, suggested by Herschel, of the presence 
of a higher kind of systematic parallax than that revealed in 
the drift of the brightest stars.* It has, however, yet to be 
discovered, and time is short for the investigations which its 
discovery would demand. 

Progress is here only possible through careful and minute 
study of the residual movements of the stars — of the move- 
ments, that is to say, which remain after the general perspec- 
tive effect of the sun's motion has been subtracted, and which 
belong, accordingly, to their individual selves. The questions 
connected with them which most immediately present them- 
selves are these : Has the sun companions on its journey, or 
does it travel alone? and. Are real stellar displacements 
governed by any obvious law ? 

The great multitude of the stars are, to all appearance, 
indifferent to the transport of our system. They have clearly 
no share in it. Just because they stand aloof, and act as 
indicators of the way, its progress becomes sensible to us. 
For motion is not alone undiscoverable, it is even unimagin- 
able without some fixed point of reference. We cannot, how- 
ever, as yet pronounce with certainty against the existence of 
a particular dynamical bond connecting the sun with some 
few of the stars, which together with it form a company asso- 
ciated by subjection to identical influences, and engaged with 
it on the same journey through space. As to the criteria by 
which such associated stars, if present, can be discriminated 
from the rest, something will be said in the next chapter. 

There, too, we will consider what answer should be given 
to our second query. A great deal depends upon it as regards 

> Phil. Trans, vol. Ixxiii. pp. 1276-7. 


our conception of the sidereal universe. Nay, the result of 
inquiry upon the point has a vital bearing upon the subject 
we have just attempted, however inadequately, to deal with. 
For the assumption that the absolute movements of the stars 
have no preference for one direction over another forms the 
basis of all investigations hitherto conducted into the trans- 
latory advance of the solar system. The Httle fabric of 
laboriously acquired knowledge regarding it at once crumbles 
if that basis has to be removed. In all investigations of the 
sun's movement, the movements of the stars have been re- 
garded as casual irregularities; should they prove to be in 
any visible degree systematic, the mode of treatment adopted 
(and there is no other at present open to us) becomes invalid, 
and its results null and void. The point is then of singular 
interest ; and the evidence bearing upon it deserves our utmost 




When the relative positions of the stars are compared at 
considerable intervals of time, they are in many cases found 
to have undergone small, but unmistakable changes of a 
seemingly capricious character. These are termed * proper 
motions,' to distinguish them from merely nominal shiftings 
due to the slow variation of the points of reference which serve 
to define the places of all the heavenly bodies as seen projected 
on the inner surface of an imaginary concave sphere. Proper 
motions are by no means easy to get at. Only from the most 
delicate observations, and with stringent precautions for 
bringing those at distant dates under precisely similar con- 
ditions, can they be elicited with satisfactory accuracy. 
Otherwise, some trifling systematic discrepancies in the com- 
pared catalogues, or accidental errors of computation, might 
pass for genuine effects of movement, with disastrous influence 
upon sidereal investigations. Hence, proper motions cannot 
generally be regarded as established unless, in addition to the 
terminal observations showing a sufficiently marked change 
of place in the course of thirty, fifty, or one hundred years, 
at least one intermediate observation is at hand to prove that 
the suspected motion has proceeded uniformly in the same 
direction, and is accordingly not the creation of personal or 
instrumental inaccuracy. 

Although not one of the millions of telescopic stars can, 
with any show of reason, be supposed at rest, less than five 
thousand of the stellar army are at present securely credited 
with measurable and progressive displacements. Many of 
these, including nearly all the ludihe of the northern hemi- 
sphere, were observed by Bradley between 1750 and 17G2 ; 


while in the southern, Lacaille's simultaneous labours serve to 
authenticate the changes of some three- score of objects to 
which he devoted especial care. So that a large stock of data 
of the required high degree of accuracy already possess an 
antiquity of one and a third centuries ; the multiplied observa- 
tions of the last sixty- years affording a further supply from 
which fresh and well-determined proper motions are con- 
tinually being, as it were, harvested from the seed planted by 
an earlier generation. 

The aspect of the heavens is, to the unaided sense, virtually 
unchanging. The constellations disclosed at the present time 
by the nightly withdrawal of the veil of twilight would be 
familiar, could they revive to survey them, to the watchers 
from the towers of Babylon. And most of the star-alignments 
given in our text-books might be as useful to students of 
celestial physiognomy a couple of thousand years hence as 
they are to-day. Every one of the indicated stars will 
indeed most probably, by that time, have shifted its place 
to the extent of many thousands of millions of miles. Yet so 
overwhelmingly vast is the sidereal scale that thousands of 
millions of miles measured upon it sink into insignificance. 

Stars advancing in a century as much as 30'^ or about -g^^^ 
of the width of the full moon, are counted rapid travellers ; and 
the swiftest class, with secular motions of IOC and upwards, 
now embraces close upon eighty members (see Appendix, 
Table IV.). Each of these, were it bright enough for casual 
perception, would in a couple of millenniums become very 
sensibly displaced even to an unskilled observer. But of the 
eighty quickest stars, less than one-half are visible to the 
naked eye, while only ten reach the fourth magnitude ; hence 
their shiftings make very little difference in the general effect 
of the starry skies. 

As might have been expected, the stars in most rapid 
apparent movement are among those nearest to the earth. 
Vicinity, in fact, and angular velocity vary together. Dis- 
placements on the sphere through identical rates of travel in 
space are large just in the proportion that the distances of the 
objects affected by them are small. Were there any approach 


to uniformity in the real velocities of the stars, we could then 
fairly estimate, from their seeming movements, their relative 
situations as regards ourselves. But there is no such 
approach to uniformity. Inexhaustible variety prevails here, 
as in every other branch of sidereal statistics. Stars with 
large proper motions are sometimes enormously, even im- 
measurably remote ; and, if stars with large parallaxes and 
little or no movement have not been discovered, it is perhaps 
because they have not yet been looked for. Their occurrence, 
for a reason to be presently explained, would be of great 
interest, and is not unlikely to be certified by measurements 
on photographic plates. 

But, however great the range of variety, it seems certain 
beforehand that, on the whole, the amount of visible motion 
in a given number of stars must decrease as their distance 
increases. And since their brightness falls off at the same 
time, although much more rapidly, there appears no escape 
from the conclusion that motion and magnitude must, on a 
wide average, vary together according to a definite ratio. 
From stars of the sixth photometric magnitude, for instance, 
we receive only one-hundredth part of the light sent to us 
by stars of the first magnitude; they must then, one with 
another, be ten times more remote. Otherwise, we should be 
driven to the unwarrantable assumption of a systematic 
diflference of real lustre between apparently large and ap- 
parently small stars. But, if the average distance of sixth- 
magnitude stars be ten times, then their mean motion should 
be only one-tenth that of stars of the first magnitude. In 
point of fact, however, this is not so. The proper movements 
of classes of stars diminish indeed very notably with their 
brilliancy, but not in the computed proportion. The dis- 
crepancy deserves attentive study. 

The average proper motion appertaining to the sixth 
magnitude, as determined directly by M. Ludwig Struve 
from 647 of Bradley's stars, is 8" in one hundred years.* 
Ten times this quantity, or 80'^ ought to be the average move- 
ment of stars of the first magnitude. But the mean derived 

* Mi^moircs dc St. Pt^tcrsbjui-gt t. xxxv. No. 3, p. 8. 


from the actually observed shiftiiigs of the twenty brightest 
stars in both hemispheres, is only 60". And since more than 
half of these transcend the standard lustre of their nominal 
rank, an excess, rather than a deficiency, of motion might 
have been anticipated. The deficiency, it is true, may prove 
to be more apparent than real. Stellar magnitudes are, in 
general, not photometrically determined, but adopted from 
Argelander's great ' Durchmusterung.' Argelander's estimates, 
however, as M. Lindemann has lately shown,* were not. 
governed by a uniform light-ratio. He unconsciously cur- 
tailed the intervals between ranks of stars above the sixth, 
and widened them for stars below the eighth magnitude. As 
the result, his sixth-magnitude stars being undervalued in 
point of light, are at a less mean distance, and possess a 
greater proper motion than if they were truly of the grade 
assigned to them. The same conclusion applies to objects of 
ranks intermediate between first and sixth ; and we thus see 
that general statements about the proper movements of star 
classes need to be received with caution. 

Yet the sluggishness of stars of the second magnitude 
seems to be a genuine fact. They should be, on the photo- 
metric scale, 6*8 times nearer to the earth than stars of the 
sixth magnitude. This would give for their mean secular 
motion 8" x 6-8 = 50''*4. Twenty-two such stars, however, 
from Bradley's and the Pulkowa catalogues, show no more 
than 1T\ And even this low figure more than doubles that 
representing the average movement of forty-two southern stars 
of 1'7 to 2-7 magnitudes, forming a descending sequence from 
the ten of first magnitude. Nor is this average improved by 
considering only the first twenty on the list, from /8 Crucis of 
1-7 to K Orionis of 2*2 magnitude. The swiftest of these 
(7 Crucis) travels only 20" a century ; taken all round, they 
move 8'', or with exactly the speed of stars presumably more 
than six times as remote ! 

The low apparent velocity of this class of stars is a very 
curious, and at present an inexplicable circumstance. It is 
ac<;entuated by the close agreement in M. Struve's results for 

* Observatory, vol. xii. p. 409. 



stars of all three ranks from the second to the fourth. A 
glance at the accompanying Table from his Memoir will serve 
better than verbal explanation to make the matter intelligible. 
The object of its compilation was to exhibit the divergence 
between the proper motions actually determined and those 
computed from the basis of the mean secular displacement 
corresponding to the sixth magnitude. In the fourth column, 
however, we have substituted figures derived from strict 
photometric star- distances for others depending upon a scale 
of distances involving precarious, even if plausible assump- 

Table of Secular Mean Proper Motionn of all Bradley's Stars differing 
by not less than eight-tenths of a magnitude. 

Mean Motion 


No. of Stars 



































It will be observed that the velocity of each order brighter 
than the sixth falls short of its theoretical amount, while that 
of the fainter orders exceeds it. We hasten to add, however, 
that (as M. Struve points out) little or no dependence can be 
placed on the above mean rate of eighth-magnitude motion 
deduced from measurements of only eleven objects. 

And now, what are we to think ? How can we account 
for the indicated deficiency of proper motion in the brighter 
stars ? Three possible explanations present themselves. It 
is conceivable that stars, say of the sixth and seventh, are 
really less effulgent bodies on an average than stars of the 
second and third magnitudes, and are consequently less 
remote than they should be on the more natural supposition 
of their equality. Their diminished distance would then at 
once render their extra celerity intelligible. Again, there 


may be a systematic increase of motion outward from the 
sun, producing in apparently small stars preponderating 
rates of displacement. Or thirdly, there may exist a special 
clafls of stars deficient in light-power, but travelling with 
exceptional speed, by the influence of which the balance of 
seeming swiftness is turned in favour of the less brilliant 
classes of stars. 

There is some probability that the last alternative represents 
at least a partial truth ; but facts might be arrayed against, 
as well as for it ; and it is, at any rate, far too soon to adopt 
it definitively. Subsidiary questions of great interest are con- 
nected with it ; as, Whether real velocity bears any relation to 
physical constitution ? Are swift stars found equally in all 
the spectral classes, or have they, on the whole, reached a 
later stage of development than more inert luminaries ? For 
satisfactory answers, however, we must wait yet some time. 

The anomalous results of Professor Eastman's recent 
inquiries* into the comparative proper motions of classified 
stars might be satisfactorily explained on the basis of our 
third alternative. Distributing 550 of the swiftest stars 
according to their brightness into nine groups, he * found an 
almost uniformly increasing proper motion as the stars grew 
fainter, until the ninth-magnitude stars were found to have a 
proper motion nearly three times as great as those of either the 
second, third, fourth, or fifth magnitudes.' * His figures are, 
nevertheless, very far from representing average movements. 
Only stars distinguished for apparent velocity were considered 
by him, and the smaller they were, the more entirely 
exceptional, obviously, their status among their equals in point 
of lustre.* Yet the disproportionate representation of faint 
objects on any list of rapidly moving stars is certainly a fact, 
and a very remarkable one. 

It comes out with strong evidence in Table V. of the Ap- 
pendix, printed by the kind permission of Professor Schon- 
feld. More than half of the seventy-six stars enumerated in it 
are invisible to the naked eye ; the three swiftest are of 6*9, 

» Bull, Phil. Soc. of Washington, vol. xi. p. 143. « Ibid. p. 167. 

■ Se ; r. Monck's remarks in Nature, vol. xli. p. 392. 

z 2 


7*5, and 8*5 magnitudes respectively; no less than fifteen 
range from the eighth to helow the ninth, while only four stars 
of the first, none of the second, and two of the third magni- 
tudes are included in the collection. The largest proper 
motion yet detected belongs to a seventh-magnitude star 
situated in the Great Bear, and numbered 1830 in Groom- 
bridge's Circumpolar Catalogue. Argelander discovered in 
1842 its pace to be such as would carry it round the entire 
sphere in 185,000 years, or in 265 over as much of it as the 
sun's diameter covers. Its annual advance, in fact, amounts 
to 7'^ ; and it has nearly equal competitors in two small 
southern stars observed by Gould during his stay at Cordoba. 
One is a 7*5 magnitude star in the Southern Fish (Lacaille 
9852), the other, one of 8-5 in the constellation Sculptor. 
Next on the roll comes 61 Cygni with a proper motion of 5""1 ; 
and a Centauri, with 8"-7, has tenth place. Twelve double stars 
are conspicuous for rapid movement, besides two wide pairs 
(bracketed together in the list), the components of which are 
severally associated by a community of swift progress. It is 
noticeable that three out of the four first-magnitude stars with 
proper motions exceeding one second of arc yearly — namely, 
Sirius, Procyon, and a Centauri — are binary combinations ; 
nevertheless, the elder Struve's general inference as to the 
quicker translation of multiple than of simple objects, has 
scarcely been borne out by further experience. The ' flying 
stars ' with which we have since made acquaintance are all 

The ' proper motions ' of stars include, as was explained 
in the last chapter, an apparent, as well as a real element ; 
they consist, in technical phraseology, of the motiis parol' 
lacticuSf optically transferred to the whole stellar multitude from 
the single real motion of the sun, and the motits pecuUaris^ 
belonging to each individual star. Means are being prepared 
of separating these two elements, at present largely blended 

But the motuB peculiaris itself is only a projection upon the 
sphere of a line of travel which may make any angle with the 
line of sight. Its conspicuousness then varies with direction, 


no less than with distance and actual velocity. A star may 
appear devoid of motion simply because the whole of it is 
* end-on ' ; while the movements of others seem large because, 
lying square to the line of sight, they are completely eflfective 
for apparent displacement. Here, just where ordinary observa- 
tion is baffled, the prismatic method comes to the rescue. The 
spectroscope * takes up the running ' for the telescope. 

But eye-observations of the motion-displacement of 
spectral lines are, as we have said,* hampered by many 
difficulties and uncertainties. Although consistently pursued 
at Greenwich during the last twelve years, no results were, or 
could be, secured concordant enough to form the groundwork 
of any extended research. Contradictions abounded ; even 
the sense of the movements registered often varied from night to 
night with the still or troubled state of the air; it could only be 
depended upon where accentuated by a rate of going somewhat 
above the average. Eapid recession was, however, fairly well 
ascertained in Aldebaran, /3 Andromedse, Begulus, and Castor, 
the velocities assigned ranging from twenty- five to fifty-eight 
miles a second. On the other hand, Arcturus, a Cygni, Vega, 
a and S Andromedae, Pollux, and a Ursae Majoris were found 
to be approaching the earth with an average speed of some- 
where about forty miles a second. 

Of these, a Cygni is, from a telescopic point of view, all but 
completely stationary, so that the whole of its motion appears 
to be directed towards our system, to which it will even- 
tually. Professor Newcomb tells us,^ become so near a neigh- 
bour as to outshine during several thousand years every 
star now visible in the sphere. The prodigious remoteness, 
however, attributed to this object by Dr. Elkin leaves open 
the possibility of its being animated by a considerable 
thwartwise movement rendered insensible by distance; and 
tends accordingly to invaUdate any conclusions as to its 
present or future course. 

Two bright stars in Orion, also virtually devoid of visible 
proper motion, prove to be rushing straight away from our 
system. The rate of recession of the briUiant Eigel, as deter- 

» See ante, p. 328. « Popular Astronomy, p. 471 (ed. 1882). 


mined by Professor Vogel in 1888,* is thirty-nine, of e Orionis, 
the middle star of the Belt, thirty-five English miles a 

A special interest and authority attach to the last-named 
results as being the earliest arrived at by photographic means. 
The prerogatives of the camera in this line of work are 
enormous. Not only do the worst mischiefs of atmospheric 
disturbance vanish with its employment, but the upshot of 
measurements executed upon one line can be checked or 
ratified by comparisons with other lines in the same spec- 
trum, and on the same plate. Where motion is in question, 
all must be equally affected by it; hence perfect security 
against illusion is afforded. The full realisation of these 
advantages through Vogel's skilful use of the spectrographic 
apparatus erected by Lim at Potsdam in 1888,* thus con- 
stitutes an advance of incalculable moment in practical stellar 

The light of the stars subjected to these delicate determi- 
nations is collected by a refracting telescope of 11^ inches 
j^ aperture, and dispersed by trans- 
''^^ "^ mission through a couple of large 
prisms. The blue ray of hydrogen 
Spectrum of Biod ^^^^^ q^ j^ ^^^^ primarily reUed 

^^dio*g^e'7uIi?^^Srspect^ ^P^n ^r measures which have their 
o* ^»«ei- zero point fixed by the simple ex- 

pedient of photographing with each star-spectrum the spec- 
trum of hydrogen in its natural position as derived from a 
Geissler tube illuminated by electricity. Relative then to this 
fiducial line, the star-line is shifted towards the blue, when 
the motion is one of approach — an effect very conspicuously 
shown by Arcturus ; towards the red, by such a movement of 
recession as Eigel exemplifies (fig. 49, from a negative). 

The more closely the data furnished from Potsdam aire 
scrutinised, the more entirely satisfactory they appear. The 
margin of possible error qualifying them was at first estimated 
at about five English miles a second ; but subsequent improve- 

» Astr. Nach. No. 2839. 

« Mottatsberichte, Berlin, March 15, 1888, p. 397. 


^a. Q^d^^*- Y,c^''-^\^ 


ments have reduced it to a much smaller quantity. A 
searching test of precision is afforded by the orbital movement 
of the earth as regards each star examined. Sometimes, 
according to the time of year, directed more or less 
obliquely towards, sometimes away from the star, it oughty 
since it is just as effective in shifting spectral Unes as the 
movement of the star itself, to produce fluctuations in the 
gross amount of displacement, while the net result, this 
known element of terrestrial motion having been eliminated, 
ought, as representing the relative movement of star and sun, 
to remain constant. The test is triumphantly sustained by 
the Potsdam determinations. A long series of measures of 
Gapella reflects with such approximate fideUty the changing 
effects of the earth's revolution, that the extreme differences 
of the results deduced for the star only slightly exceed four 
EngUsh miles a second.^ It is scarcely possible that the con- 
cluded rate of withdrawal of sixteen miles a second can be 
wrong either way by as much as two miles. The much 
smaller approaching movements of a Persei and Procyon 
(seven miles a second in each case) were arrived at with still 
more satisfactory accuracy. The outcome of each of the two 
pairs of measures executed, agreed within less than half a 
mile ! A marginal uncertainty of one mile and a half appears to 
attach to the rate of approach of sixteen miles a second attri- 
buted to the pole-star ; and it diminishes to about one mile for 
Aldebaran's swift recession of thirty miles a second. At last 
then, firm ground has been reached in these critical investi- 
gations, the data suppUed by which may now be fearlessly em- 
ployed for the solution of many outstanding problems. That, 
for instance, of the variable movement oi Sirius, found at 
Greenwich to alternate recession and approach with indi- 
cations of cyclical regularity in a period of several years. The 
reality of the phenomenon (upon which the most recent ex- 
perience of the Greenwich observers has thrown some doubt) 
will now before long be established or decisively disproved. 

To the Table of Parallaxes included in the Appendix, a 
column has been added showing the linear velocities across the 

» Astr. Nach. Nos. 2896-7. 


visual ray of the fifty-two starB at known distances. They 
range from nearly four hundred to one and a half miles, and 
average sixty miles a second. And this for only one component 
of their motion ! Nor is any reason imaginable why the 
other component — that in the line of sight — should be inferior 
n amount, even though Yogel's results, so far as they have 
gone, give a much lower mean rate. Later on, perhaps, the 
spectrographic method may bring to light radial velocities as 
extraordinary as certain tangential ones, by which the average 
of thwartwise movement is forced up to the above high 

The first of these startling examples to become known was 
\in 1880 Groombridge. The large proper motion and small 
I parallax of this star compel the ascription to it of a speed, 
taking into account only that part of it lying square to our 
view, of at least two hundred miles a second — a speed uncon- 
trollable, according to Professor Newcomb, by the combined 
attractive power of the entire sidereal universe. For his calcu- 
lations show that the maximum velocity attainable by a body 
falling from infinity towards and through a system composed 
of 100,000,000 orbs, each five times as massive as our sun, 
and distributed over a disc-like space 80,000 light-years in 
extent, would be twenty-five miles a second.^ But 1880 Groom- 
bridge possesses fully eight times this speed; and because 
velocity varies with the square root of the attracting mass, a 
world of stars of more than sixty-four fold the potency of 
that assumed as probable, would be required to set this object 
moving as it does unquestionably move ! 

Now the velocity producible by an attractive system is the 
limit of the velocity it can control — that is, bend into a closed 
curve. It is then certain that unless the stellar system 
possesses what we may call occult gravitational energies, the 
star in question cannot be one of its permanent members. 
Virtually in a straight line and without slackening, it will 
pursue its course right across the starry stratum it entered 
ages ago on its unknown errand, and will quit ages hence to 
be swallowed up in the dusky void beyond. There is, how- 

* Pop. Astronomy t p. 499. 


ever, an alternative suppoBition. The star may be acted upon 
by unknown compulsive influences. 

This gains increased probability as the number of abnor- 
mally swift stars multiplies. Groombridge 1880 is no longer 
the only ' rimaway ' of our acquaintance. Linear stellar speed, 
apart from that share of it directed along the line of sight, 
exceeds in fact the computed maximum of twenty-flve miles a 
second in nearly half the cases in which it has been ascertained, 
and the excess is here and there enormous. Thus, Arcturus 

* pioves palpably through heaven ' at the rate of 876 miles a 
second, and the velocity of /* CassiopeisB is but slightly in- 
ferior, Groombridge 1830 standing only third in point of real 
celerity. Next to it comes f Toucani with its 101 miles, and 
four southern stars besides progress at above sixty miles per 

' Flying stars ' can then no longer be regarded as mere 
intruders into stellar society. Whether or not belonging to it 

* for better for worse,* they evidently at present form an im- 
portant part of it, and the problem they present cannot be 
excluded from a general consideration of sidereal mechanism. 
Indeed, they furnish a most significant index to the workings 
of its secret springs. They pursue their careers, so far as 
observation can yet tell, in right lines, and at a uniform speed. 
Their high velocities would be otherwise less perplexing ; for 
they might plausibly be supposed due to the powerful attrac- 
tion of invisible bodies in their neighbourhood, and to represent, 
by analogy, the rush past the sun of highly eccentric comets. 
But the evidence is wholly against any such hypothesis. All 
proper motions known to us — whether of single stars, or of the 
centres of gravity of multiple stars — are sensibly rectilinear. 
The centres of curvature, presumably, of the imaginary lines 
traced out by them are inconceivably remote. A straight 
line is only part of the circumference of a circle of infinite 

The fact then confronts us that not a few of the stars 
possess velocities transcending the power of government of the 
visible sidereal system. Is that system then threatened with 
dissolution, or must we suppose the chief part of its attractive 


energy to reside in bodies unseen, because destitute of the 
faculty of luminous radiation ? No answer is possible ; con- 
jecture is futile. We are only sure that what we can feebly 
trace is but a part of a mighty whole, and that on every side 
our imperfect knowledge is compassed about by the mystery 
of the Infinite. 

Physical peculiarities are not, in any obvious way, related 
to these excessively rapid movements. Axcturus, although, 
as regards the character of its spectrum, a * solar ' star, com- 
bines, we have found reason to believe, a higher temperature 
with stronger absorption than are present in the sun. Its 
mass is probably enormous ; its light-power certainly is. It 
proves, in fact, by a moderate estimate, to be equivalent to 
that of eight thousand suns ! The next swiftest star, fi Gassio- 
peisB, appears to be about nineteen times more luminous, 1880 
Groombridge to be slightly less luminous than the sun. Neither 
of these stars shows a very distinctive spectrum. Stars with 
banded and gaseous spectra, and variables of all classes, are, 
as a rule, almost destitute of proper motion ; but this may be 
an effect of remoteness, rather than of genuine inertness. 

An unmistakable connection, however, exists between 
proper motion and sidereal locality. The late Mr. Proctor 
drew attention to the prevalence in certain regions of the sky 
of what he termed * star-drift.' Here and there, unanimity of 
movement is, to some extent, substituted for the caprice which 
is the most striking superficial feature of stellar displacements, 
when deduction has been made of then* common perspective 
element. Amid seeming confusion, order and purpose by 
glimpses reveal themselves. Battalions of stars— 'flying 
synods of worlds ' — regardless, as it were, of the erratic flittings 
of the casual surrounding crowd, march in widely extended 
ranks, by a concerted plan along a prescribed track, under 
orders sealed perhaps for ever to human intelligence. 

Among the stars situated between Aldebaran and the 
Pleiades, there is next to no relative movement. They all 
drift in company towards the east, by about 10'' in a century. 
All, we mean, that have been investigated. It has not yet 
been ascertained to what extent the fainter neighbouring stars 


share a tendency largely, if not wholly, due to the son's oppo- 
sitely directed progress. 

Five of the * Seven Stars* {septem triones) forming the Plough 
(those excluded being the ' Pointer ' next the pole, and 17 at 
the extremity of the handle) were regarded by Mr. Proctor as 
members of a vast united group advancing, but at a higher speed, 
towards the same point with the sun. But his inference has 
been in a measure overthrown by Dr. Auwers's exact deter- 
minations of the very small proper motions affecting these 
brilliant objects. The concert assumed to exist between them 
is thus so gravely disturbed that only two out of the five can 
now be safely bracketed as companions. These are s and ? in 
the Plough-handle, the separation of which by 4° 22' of a great 
circle implies a gap so vast as to be measured by many — we 
cannot tell how many — years of Ught-travel. The system 
thus constituted is, at the very least, a quintuple one, since 
f UrsHB, as our readers are aware,' carries with it three 
dependent stars, one (Alcor) visible to the naked eye, one a 
telescopic attendant, the third revealed only by the spectro- 
scopic effects of orbital motion. 

It is scarcely likely, however, that the system is complete 
in itself. Common proper motion does not necessarily imply 
the mutual revolution of the objects to which it belongs. 
What it does imply is their systemic connection. But that 
connection need not be of the kind exemplified close at hemd 
by the earth and moon. It may rather be such as prevails 
between the earth and Venus, or between Jupiter and Saturn. 
The group in Ursa Major, it is safe to assert, includes 
examples of both kinds of relationship. Of the movements 
of two satellites, Mizar (f Ursa?) is the undoubted mainspring. 
The status of Alcor is dubious. Its path at present appears 
strictly rectilinear; but curvature relative to the large ad- 
jacent star may be rendered sensible in it by future observa- 
tions. About what might be called the personal independence, 
however, of the distant s, there is Uttle room for doubt. 
Although dominated by the same influence, it advances on its 
own account ; and since the probabilities are strong against 

• See ante, p. 212. 


its keeping strict pace with its travelling companion, their 
relationship will perhaps eventually cease to be traceable. 
Slight inequalities betraying differences in the period of 
revolution round the same remote centre may easily co-exist 
with what is known as common proper motion. Such dis- 
crepancies can alone hold the stars affected by them aloof 
from binary combination. While travelling along parallel 
lines, they have still a relative velocity exceeding, at their 
distance apart, the power of their mutual gravitation to sway 
into an ellipse. One must hence fall very slowly behind the 
other, as Saturn falls behind Jupiter after conjunction. 
Evidence of their affinity is then only temporarily accessible 
to us. After many ages it will cease to come within the 
sphere of our possible recognition. There may be, probably 
are, in distant parts of the sky, stars revolving in boundlessly 
spacious orbits round the same focus of attraction with 
e and f Ursae ; but we have no means of identifying them. 

'Partial systems,' governed presumably from without, 
are of tolerably frequent occurrence. The first to become 
known was discovered by Bessel in 1818.* It is composed of 
a fifth and a seventh magnitude star known respectively as 
86 A Ophiuchi and 30 Scorpii, more than twelve minutes of arc 
apart, yet endowed with an accordant movement of l"-22 
yearly. The former star has a close attendant; and an 
intermediate minute object also forms part of the company.* 
Another interesting quadruple group was detected by Flam- 
marion in 1877.^ Two couples in the Swan, one revolv- 
ing, the other in appearance fixed, separated by em interval 
of 15', drift slowly southward together in a direction nearly 
perpendicular to the line of march of the sun. Their move- 
ment is hence ' proper ' to themselves, perspective effects being 
unconcerned with it. The stationary pair is the fifth-mag- 
nitude yellow star, 17 Cygni, with its bluish satellite at 26" ; 
the circulating pair consists of two eighth-magnitude stars 
at 3'', numbered 2576 in Struve's great Catalogue. 

* Fundamenta AstronomicB^ p. 311. 

' Flammarion, Comjptes Bendust t. Ixxxv. p. 783 ; Cat, des itoUes Doubles^ 
p 105. 

C Bendus. t. Ixxzv. p. 510. 


The most curious instance of concerted movement yet 
brought to light is afforded by two ninth-magnitude stars in 
Libra, discovered by Schonfeld in 1881 to progress across the 
sphere at the exceptionally quick rate of 8"'7 annually.* 
Notwithstanding the wide interval (50 separating them, their 
advance seems perfectly harmonious. They flit side by side, 
as if rigidly connected across a chasm probably some thousands 
of millions of miles in width. The measures of their parallax, 
now progressing at the Cape, will soon decide whether they 
are as near the earth as might be supposed, and supply the 
means of determining their true velocity, so far as it lies 
square to the line of sight. 

To the question — Has the sun any associatesin his journey 
through space ? only a provisional answer can as yet be given. 
None are known, but investigations on the point are barely 
nascent. The peculiarities which we should expect beforehand 
to attend such companion-stars are comparative proximity 
and relative immobility. They should have sensible parallaxes, 
and be devoid both of radial and tangential velocity. Neither 
spectroscopic nor telescopic evidence of motion should be 
derivable from them. It is certain, indeed, that no star up 
to this completely investigated combines these characters ; but 
then they could not possibly be found in the * proper motion ' 
stars chosen by preference as the subjects of parallactic 
observations. When more has been done in photographically 
registering line-of-sight movements, stars may perhaps be 
discovered sensibly fixed as regards the sun, because borne 
along with him at the same translatory speed. The con- 
struction of such a group, and the investigation of the 
peculiarities of its members, might open up a fascinating 
branch of inquiry. But its existence is purely imaginary; 
not a single star can be pointed, to as at all likely to be 
coupled with the sun in his advance. They should be 
diligently searched for, since their occurrence, or non-occur- 
rence, must be of essential importance to any theory of 
sidereal construction ; yet without undue confidence as to the 
result of the search. 

> Siteungsherichte Niedcrrheinischer Oes, Bonn ISSl, p. 172. 


If the system formed by the stars be destined to perma- 
nence in its present shape, some general law of movement 
must be obeyed by them. Even if its state be one of pro- 
gressive modification, a prevalent method of change ought to 
make itself felt. Local irregularities, however, so eflfectually 
disguise the fundamental harmony that its presence may long 
continue a matter of speculative belief. 

The assumption is indeed indispensable, as Dr. Schonfeld 
pointed out in 1888,' that the motions of the stars are some- 
how related to the plane in which the vast majority of them 
are disposed. For otherwise their actual configuration would 
be a wildly improbable accident of the time in which we live. 
The Milky Way, to put it otherwise, should be regarded as 
an evanescent phenomenon, unsustained by any persistently 
acting forces, the outcome of a hundred millions of casual 
conjunctions. If this be incredible (as it surely is), then we 
are constrained to admit a preference, in the long run, among 
stellar displacements for the grand level of stellar aggregation. 
The Milky Way must be, in some true sense, what Lambert 
called it in the last century — the * ecliptic of the stars.' 

Sir John Herschel imagined the law of harmony to consist 
in a general parallelism of stellar motions, involving a kind 
of systematic circulation, as of a solid body round an axis 
perpendicular to the galactic plane. Innumerable exceptions 
to any such rule are of course to be found, but they were 
assumed, in the upshot, to be mutually destructive, the main 
* stream of tendency ' flowing on irrespective of them. But 
it is difficult to conceive a physical basis for a quasi-rotational 
system wholly without warrant from experience. More 
plausible is M. Ludwig Struve's view that the main part of 
the revolutions of the stars roimd their common centre of 
gravity situated in the Milky Way, are performed in planes 
slightly inclined to that of the zone towards which they are 
concentrated.^ His attempt, indeed, to elicit a ' rotation- 
component ' from the secular movements of Bradley's stars, 
proved unavailing. Yet this is not decisive against the truth 

* V, J. S. Asir. Oes. Jahrg. xvii. p. 255. 

» Zlimoires de St, Pitersbourg, t. xxxv. No. 3 pp. 5, 19. 


of an hypothesis which leaves open the possibility of a balanced 
stellar circulation pursued in opposite senses. 

M. Bancken was in 1882 more fortunate with a selected 
list of stars.' He admitted only those within thirty degrees 
on either side of the Milky Way, and possessed of annual 
proper motions not exceeding a quarter of a second. The 
solution of his equations showed these movements to include 
a common element of very slow progressive increase of 
galactic longitude. That is to say, the 106 stars considered 
were found subject to a drift along the Milky Way in the 
direction from Aquila upward towards Cygnus and Cassiopeia, 
and down past Capella through the Club of Orion towards the 
Ship. The reality and extent of this drift will be a matter for 
future investigation. Should the one be confirmed, and the 
other ascertained, something like a clue to the labyrinth of 
stellar movements will have been provided. Even the sugges- 
tion of its presence is useful as an indication of what may be 
looked for with some chance of success. 

> Astr. Nach, No. 2482. 




The Milky Way shows to the naked eye as a vast, zoue^ 
shaped nehula ; bat is resolved, with very slight optical assist- 
ance, into innamerable small stars. Its stellar constitution, 
already conjectured by Democritus, was, in fact, one of 
Galileo's earliest telescopic discoveries. The general course 
of the formation, however, can only be traced through the 
perception of the cloudy effect impaired by the application 
even of an opera-glass. Rendered the more arduous by this 
very circumstance, its detailed study demands exceptional 
eyesight, improved by assiduous practice in catching fine 
gradations of light. Our situation, too, in the galactic plane 
is the most disadvantageous possible for purposes of survey. 
Groups behind groups, systems upon systems, streams, sheets, 
lines, knots of stars, indefinitely far apart in space, may all 
be projected without distinction upon the same sky-ground. 
Unawares, our visual ray sounds endless depths, and brings 
back only simultaneous information about the successive 
objects met with. We are thus presented with a flat picture 
totally devoid of perspective-indications. Only by a long 
series of inductions (if at all) can we hope to arrange the 
features of the landscape according to their proper relations. 
To the uncritical imagination, the Milky Way represents a 
sort of glorified track through the skies — 

A broad and ample road, whose dust is gold 
And pavement stars, as stars to thee appear, 
Seen in the galaxy, that milky way, 
Which nightly as a circling zone thou seest, 
Powdered with stars. 


■ In American-Indian fancy a mysterious * path of souls,' its 
popular German name, ^ die Jakobsstrasse/ recalls the time 
when it stood as a celestial figure of the way of pilgrimage 
to Compostella, This superficial impression is, however, 
effaced by closer inspection, and an aspect is assumed re- 
sembling rather that of a rugged trunk marked by strange 
cavities and excrescences, and sending out branches in all 

The medial line of the Galaxy is sensibly a great circle.* 
This shows that the formation extends symmetrically on all 
sides of us. We are placed somewhere in its main level, 
though not necessarily very near its centre. The superior 
brilliancy of the Milky Way in its southern sections has, 
indeed, suggested that the sun lies much nearer to the edge of 
the apparent ring there than elsewhere ; but a positive con- 
clusion on the point would still be premature. None of the 
various movements affecting the earth bears any obvious 
relation to the starry collection around it. Neither the 
equator, the ecliptic, nor the line of the sun's way, shows any 
trace of conformity to its plane. The great circle of the Galaxy 
is inclined about sixty-three degrees to the celestial equator, 
which it intersects in the constellations Monoceros and Aquila. 
It passes in Cassiopeia within twenty-seven degrees of the 
north pole of the heavens, in Crux, as near to the south pole, 
while its own poles are located respectively in Coma Berenices ^ 
and Cetus. Over two-thirds of the celestial circuit, the 
general unity of this stupendous structure is preserved. 
Broken, however, near a Centauri by the interposition of a 
great fissure, it is only regained, after an interval of some 
120^, through the reunion, in the neighbourhood of e Cygni, 
of the separated portions. Involuntarily, the image presents 
itself of a great river, forced by an encounter with a powerful 
obstacle to throw its waters into a double channel, lower down 
merged again into one. The intervening long strip of islanded 
rock and gravel might stand for the great rift between the 
branches of the sidereal stratum, which, although to the eye, 

* Gould, Uranometria Argentina, p. 370. 

« R,A. 12h. 41m. 20s., Deo. + 227°21', according to Gould. 

A A 


owing to the effect of contrast with the ' candid way ' on either 
side, darker than the general sky, is in reality nowhere quite 
free from nebulous glimmerings. It is encroached upon by 
fringes, effusions, and filaments, spanned by bridges of light, 
and here and there it is half filled up by long, narrow, dis- 
connected masses, or pools of nebulte, lying parallel to the 
general flow of the stream. One such ' brilliant and tortuous 
streak ' ^ extends, in almost complete isolation, oyer nearly 
20^, from the tail of Serpens across a corner of the Shield of 
Sobieski. Moreover, only one of the two principal branches 
— that traversing Aquila and the bow of Sagittarius — is con- 
tinuous. The other, after covering the tail of Scorpio * with 
a complicated system of interlaced streaks and masses,' ^ 
dies out in Ophiuchus, about fifteen degrees south-west of 
the termination, just at the equator, of the arm sent out to 
meet it through Cygnus. The gap is, however, partially 
veiled by a faint luminous extension from the south, and 
thus shows as absolute only over some five degrees of the 

This is not the only interruption to the course of the Milky 
Way. And that occurring in Argo is the more remarkable 
that it cuts sheer across the entire, undivided stream. Here, 
at south declination 83°, the formation, Sir John Herschel 
says, * opens out into a wide, fan-like expanse, nearly 20° in 
breadth, formed of interlacing branches, all which terminate 
abruptly in a line drawn nearly through X and 7 Argus.'* 
On the opposite, or eastern side of a moderately broad blank 
space, a similar assemblage of branches converges upon the 
variable star tj Argus. There is an obvious correlation of 
structure on either side of the chasm ; subdivisions mutually 
correspond ; the broken series on one margin is resumed on 
the other. 

The impression is strongly conveyed that star-strata once 
united, have here yielded to the influence of some unknown 
dispersive force or forces, perhaps still in operation. Yet we 

> Vran, Argent p, 381. « Hergchel, OtUUnes, art. 789. 

■ Ibid. art. 787. 


can scarcely hope ever to command the means of testing the 
conjecture. For the proper motions of the faint telescopic 
stars near the edges of the gap are no doubt of such excessive 
seeming minuteness that centuries, nay, millenniums may pass 
before they can become perceptible. 

The representation of the Milky Way as a uniform starry 
stream is purely conventional. Its real texture is of a curdled 
or flaky description.' Between Perseus and Sagittarhis, Sir 
William Herschel counted eighteen luminous patches, ' re- 
sembling the telescopic appearance of large, easily resolved 
nebulfiB ; ' * and his son perceived the lucid ramifications in 
Sagittarius to be made up of ' great cirrous masses and 
streaks,' the appearance, as his telescope moved, being ' that of 
clouds passing in a scud, as the sailors call it.' Further on, 
he remarks : * the Milky Way is like sand, not strewed evenly 
as with a sieve, but as if flung down by handfuls (and both 
hands at once), leaving dark intervals, and all consisting of 
14th, 16th, and 20th magnitudes,^ down to nebulosity, in a 
most astonishing manner.'^ 

The bright spaces of the galactic zone are commonly sur- 
rounded and set off by dark winding channels, and the rapid 
alternation of amazingly rich with poor, or almost vacant 
patches of sky, is a constantly recurring phenomenon.* The 
most remarkable instance occurs in the Southern Gross, the 
brilliant gems of which emblazon a broad galactic mass very 
singularly interrupted by a pear-shaped black opening eight 
degrees long by five wide, named by early navigators the 
* Coal-sack.' This yawning excavation, which is, however, not 
absolutely denuded of stars, figures in Australian folk-lore as 
the embodiment of evil in the shape of an Emu, who lies in 
wait at the foot of a tree, represented by the stars of the Cross, 
for an opossum driven by his persecutions to take refuge 

* Houzeau, UranonUtrie OirUrale, p. 16, 1878 ; Klein, Wochensckrift fUr 
Astr. 1867, p. 288. 

* Phil. Trans, vol. civ. p. 282. 

' HerBoheVs * 20th magnitude ' corresponds approximately with the 14th on 
the photometric scale. 

* Cape Observations, p. 388. 

* Herschel, Outlines, arts. 790, 797. 

A A 2 

356 The system of the staiis 

among its branches.' The legend reads almost like a Chris- 
tian parable. 

Partial galactic vacuities, evidently of the same nature 
with the southern Goal-sack, though less perfectly developed, 
occur elsewhere, notably in Cygnus ; but they are incon- 
spicuous to casual observers. A telescopic perforation has 
been adverted to by Mr. Barnard.^ 'A most remarkable 
small, inky black hole in a crowded part of the Milky Way ' 
in Sagittarius, is described by him as ' about 2' in diameter, 
slightly triangular, with a bright orange star on its north 
preceding (north-westerly) border, and a beautiful little cluster 
following.' This singularly abrupt interruption to the clouds 
and * breaking sprays ' of stars around it, appears clearly on 
a negative taken in 8h. 7m., August 1, 1889, with highly 
suggestive result as regards the same observer's project for 
photographically charting the Milky Way. 

Meantime, incomparably the best delineation of the sections 
fo it visible in the northern hemisphere, has recently been com- 
pleted at Parsonstown after five years of labour, by Dr. Otto 
Boeddicker. The amount of unmistakably genuine detail re- 
corded in it, despite climatic conditions of the least propitious 
sort, is simply astonishing ; and the peculiarities thus brought 
to light are in many cases of a novel and significant charac- 
ter. The general effect may be best described as that of a 
thick stem of light, closely set with curvilinear ramifications ; 
the stem itself being riddled with dusky convolutions, intricate 
passages, and ' horse-shoe ' or * key- hole ' apertures, separated 
by lustrous wisps and nebulous * pointed arches.' The circum- 
stance that * feelers are thrown out towards nebulae and 
clusters,' ^ is of profound interest. The Andromeda nebula, 
for instance, terminates a feeble branch starting from a Cas- 
siopeiae. The Pleiades stand at the peaked summit of a dim 
arch springing, on one side, from near /8 Tauri, on the other, 
from € AurigflB. The Hyades are separately involved ; Preesepe 
is all but reached by a long streamer issuing from the vicinity 
of 13 Canis Minoris ; while a thin sinuous effusion, perhaps of 

» MacPherson, Joum. R. Soc. N. S, Wales, 1881, p. 72. 
» Monthly Notices, vol. 1. p. 811. ■ Ibid. p. 12. 


a spiral nature, Includes the great nebula in its sweep through 
Orion. A parallel streak of barely traceable nebulosity (also 
noticed by Mr. Gore in India) obviously guides or is guided 
by a flow of small, but still lucid stars, marking the lion's 
skin supposed thrown over the left shoulder of the haro ; but 
other instances of apparent relationship between galactic wisps 
and bright stars are not improbably of purely physiological 
origin. A broad effusion is depicted by Dr.Boeddicker as just 
enveloping the pole-star, and all the cavities ai;d interstices 
of the formation appeared to him filled with dim luminosity. 

The Milky Way is throughout subject to great and sudden 
variations of width. A brilliant stream, no more than three 
or four degrees in extent where it enters the Cross, it spreads, 
exclusive of faint borders, to twenty-two degrees in its bifur- 
cated section between Ophiuchus and Antinous. At some 
spots, too, the nebulous effect to the eye fades away imper- 
ceptibly along the margins ; ^ at others, the line of demarca- 
tion is so sharp that a telescope may have one half of its field 
crowded with galactic stars, while the other half is well-nigh 
blank.' A definite semi-circular boundary, for instance, limits 
the formation near ^ Aquilse ; its southern edge in Ophiuchus 
was remarked by Sir John Herschel, as ' terminated by an 
irregular nebulous fringe as if lacerated ; ' ' and marginal 
projections, knobs, and bristling outliers can, in other parts, 
easily be traced. The prevalent rule seems to be that the 
smaller the stars considered, the more abrupt is the com- 
mencement of the Milky Way ; while a more and more gradual 
condensation accompanies each step upward in brightness.^ 

Sir Wilham Herschel was perfectly satisfied that, with his 
twenty-foot reflector (equivalent to a modern refractor about 
fourteen inches in aperture), the Milky Way was, in general, 
' fathomable.' The stars composing it, that is to say, were of 
definite numbers, and appeared projected upon a perfectly 
black sky. But this was not so everywhere ; certain points 
completely baffled the penetrative faculty of his instrument. 

» W. Herschel, PhU. Trans, vol. dv. p. 283. 

* Proctor, Ufwverse of Stars^ p. 86. 

' Klein, Wochemchrift, 1867, p. 385 ; J. Herschel, Cape Results, p. 889. 

^ Celoria; Menwrie del, B, Istitwto LomlxirdOf t. xiv. p. 837, 


One such was met with in Cepheus, where he found the small 
stars to become ' graduaUy less till they escape the eye, so that 
appearances here favour the idea of a succeeding, more distant 
clustering part.' And he remarked, in exploring between 
Sagitta and Aquila, that ' the end of the stratum cannot be 
seen.' * Again, in the galactic branch traversing Ophiuchus, 
Sir John Herschel encountered ' large milky nebulous irregular 
patches and banks, with few stars of visible magnitudes;' 
described a.* very large space ' of the Milky Way in Sagittarius 
as ' completely nebulous like the diffused nebulosity of the 
Magellanic Cloud ; ' * and observed a similar spot in Scorpio, 
* where, through the hollows and deep recesses of its compli- 
cated structure, we behold what has all the appearance of a 
wide and indej&nitely prolonged area strewed over with dis- 
continuous masses and clouds of stars which the telescope at 
length refuses to analyse.' ' 

Even with the best telescopes of recent construction, this 
perplexing and indeterminate aspect cannot be altogether got 
rid of. Professor Holden tells us that the thirty-six inch Lick 
achromatic shows in the Milky Way ' no final resolution of its 
finer parts into stars. There is always the background of un- 
resolved nebulosity on which hundreds and thousands of stars 
are studded — each a bright, sharp, separate point.'* The 
stellar nature of the lingering nebulosity is strongly indicated ; 
but it cannot be pronounced off-hand whether it is due to the 
presence of innumerable small stars mixed up in the same 
region with larger ones, or to the indefinite extension outward 
into space of galactic agglomerations. 

The explanations attempted of these complicated phenomena 
may be divided into disc-theories, ring-theories, and spiral- 
theories. The * disc-theory ' of the Milky Way was first pro- 
pounded by Thomas Wright, of Durham, in 1750. He supposed 
all the stars to be distributed in a comparatively shallow layer, 
producing, by its enormous lateral extension, the effect of 
annular accumulation. Irregularities, he thought, were partly 

* PhU, Trans, vol. ovii. p. 326. * Cape Observations^ p. 889. 

' OuOines of Astr art. 798. « fifid. Afeu.^oig. 1888, p. 298. 


due to our eccentric position within the stratum, partly to 
'the diversity of motion that may naturally be conceived 
amongst the stars themselves, which may, here and there, 
in different parts of the heavens, occasion a cloudy knot of 
stars.' * 

To this view Sir William Herschel gave wide currency and 
apparent stability by the application to its support of his 
ingenious method of 'star-gauges.' By counting the stars 
simultaneously visible in his great reflector in various portions 
of the sky, he showed that their paucity or abundance depended 
upon the situation of the gauge-fields relative to the Milky 
Way. In its neighbourhood, stars were copiously — far from it, 
they were sparsely — distributed. And this by a regular pro- 
gression of density from the galactic poles to thQ galactic 
equator, the latter region being on an average thirty times 
richer than the former. Now, if we were to admit, as Herschel 
did, a nearly equable scattering of stars, there would be no 
alternative but to suppose the sidereal system extended in any 
direction proportionately to the number of stars seen in that 
direction. Their crowding should, on that hypothesis, be purely 
optical — the effect of the indefinite spreading out in the line 
of sight of their evenly serried ranks. Sounding the star- 
depths upon this principle, Herschel measured the length of 
his line by their seeming populousness, and constructed, from 
the numerical data thus obtained, the 'cloven disc' model, 
long accepted as representing the true form of the stellar 

But his own observations at the very moment of enouncing 
this theory, fatally undermined it. Already in 1785, he re- 
marked that two or three hundred * beginning, or gathering 
clusters,' might be pointed out in the galactic system, and he 
surmised its eventual separation, ' after numbers of ages,' into 
so many distinct 'nebulsB.'^ 'Equable scattering,' then, was 
an ideal state of things long since abolished by the * ravages 
of time.' The conviction that such was the case grew with 
his experience. 'The immense starry aggregation of the 

- An Original Theory of the Universe, p. 63, 
* PhU. Trans, vol. hczv. p. 355, 


Milky Way, he wrote in 1802/ 'is by no means unifonn. 
The stars of which it is composed are very unequally scattered, 
and show evident marks of clustering together into many 
separate allotments.' Nor did he fail to perceive, from the 
gradual increase of brightness towards the centres of these 
' allotments/ that they tended to assume a spherical form, 
and thus suggested ' the breaking-up of the Milky Way, in all 
its minute parts, as the unavoidable consequence of the cluster- 
ing power arising out of those preponderating attractions 
which have been shown to be everywhere existing in its com- 
pass.' 2 The formal announcement of his conviction ' that the 
Milky Way itself consists of stars very differently scattered 
from those which are immediately about us,' ^ amounted to 
a recantation of the principle of star-gauging. 

With it disappeared from Herschel's mind the conception 
of an optically-produced galaxy. In his ultimate opinion, the 
actual corresponded very closely with the apparent structure : 
it was composed, that is to say, mainly, if not wholly, of real 
clouds of stars. Credit was thus restored to the early im- 
pression of Galileo, who in 1610 described the Milky Way as 
* nothing else but amssa of innumerable stars planted together 
in clusters.' * 

Wilhelm Struve's* effort towards the re-organisation of 
the stratum-theory, though aided by all the resources of his 
great ability and address, could scarcely be counted as a step 
in advance. Substituting for the hypothesis of equable dis< 
tribution that of concentration in parallel planes, he imagined 
the average interval of space between the stars to diminish 
regularly with approach to the central horizon of the system. 
The swarming aspect of the Milky Way was hence regarded 
as agreeing with fact, but the annular appearance as being 
illusory. Of illimitable dimensions, the system was conceived 
to stretch away, still preserving its specific character, to an 
infinite, or at least unimaginable remoteness, comparatively 

» PM. Trans, vol. xoii. p. 495. 

« Ibid, vol. civ. p. 282. ■ Ibid. vol. xoii. p. 480. 

* Sidereus Ntmcvust trans, by E. B. Oarlos, p. 42, 

* Etudes d^Astronomie SUllaire^ 1847. 


narrow vimal bounds being, however, set to it by a supposed 
extinction of light. 

But the quasi-geometrical regularity of Struve's galaxy is 
belied by innumerable details of the original. The swell of the 
tide of stars towards the galactic plane is neither uniformly 
progressive,* nor does it proceed without conspicuous inter- 
ruptions. Thus, the region near the horns of Taurus, although 
close to the Milky Way, is absolutely the poorest in the 
northern hemisphere ; ^ and an almost clean-swept space in 
Scorpio, on meeting which Sir William Herschel exclaimed in 
amazement, ' Hier ist wahrhaftig ein Loch im Himmel ! ' lies 
on the verge of the galactic stream. But it is the openings in the 
formation itself which afford the most decisive evidence against 
any modification of the stratum-theory. Is it credible that a 
boundlessly extended layer of stars should be pierced, in many 
of its densest portions, by tunnels converging directly upon 
our situation within it ! ^ No sane mind, we venture to say, 
realising all that such an assertion implies, can assent to it. 
But, indeed, the entire conformation of the Milky Way, — its 
streaming offsets, convoluted windings, promontories, and 
sharply bounded inlets no less than the breaches of its con- 
tinuity — all flatly contradict the view of its being the optical 
creation of any universally valid law of star distribution. 

We seem then led to the alternative belief that it is a de- 
finite structure, at a definite distance from ourselves — a belief 
forced upon Sir John Herschel by his Cape experiences, not- 
withstanding his natural reluctance to drift far away from the 
position originally taken up by his father. The shape sug- 
gested by him for the galaxy was that of ' a flat ring, or some 
other re-entering form of immense and irregular breadth and 
thickness.' ^ Expanded indefinitely along the central plane, 
the new model scarcely differed from the old except in so far 
as the idea of homogeneous construction was given up. The 
disc remained, but with its centre scooped out. The solar 
system was located in an enormous space of relative vacuity. 

> C. S. Peiroe, Harvard AnnaU, vol. ix. p. 174. 

' Argelander, Bormer Beob. Bd. v. EinleUung ; Proctor, Univerae of Stars, 
p. 82. 

" Prootor, loc eit. p. 15. * OiUU/nes, art. 788. 


The Milky Way was then supposed to consist of an inde- 
finite number of stellar collections * brought by projection into 
nearly the same visual line ' — to represent the foreshortened 
effect (more especially at a particular spot in Sagittarius) of 
* a vast and illimitable area scattered over with discontinuous 
masses and aggregates of stars in the manner of the cumuli 
of a mackerel-sky.' ^ But in an assemblage of this nature — 
seen edgewise — a ' Goal-sack ' would be a phenomenon as 
anomalous as in a uniform stratum ; nor could it, without 
violent improbability, be conceived of as rent by the colossal 
fractures dividing the actual Milky Way in Argo and Ophiu- 

To remedy these inconveniences, Professor Stephen Alex- 
ander devised in 1852,^ upon the model of the wheel-shaped 
nebula in Virgo (M 99), a spiral galaxy with four curvilinear 
branches diverging from a central cluster formed by the sun 
.and lucid stars. By properly adjusting the mode of projection 
.of these radiating star-streams, the effects of rifts and coal- 
sacks were duly produced ; but the arrangement, however 
admired for ingenuity, gave no persuasion of reaUty, and 
quickly dropped out of remembrance. Essentially different, 
although with some features in common, was that by which 
Mr. Proctor replaced it in 1869.* Rather than a 'spiral,' 
indeed, the new design resembled a bent and broken ring, with 
long, riband-like ends, looped back on either side of an opening, 
accommodated to the shape of the gap in the visible structure 
in Argo. One of these loops, by the apparent inter-crossing 
of its near with its remoter branch, was supposed to generate 
the Coal-sack in Crux ; while the other end, trailing lengthily 
backward, afforded a deceptive effect of bifurcation. Excessive 
distance was brought in, as in Professor Alexander's scheme, 
to explain the cessation of nebulous light in Ophiuchus. 

Of the manifold objections to which this hypothesis is 
liabla,* only two need here be mentioned. In the first place, 
it involves a wholly inadmissible rationale of the openings seen 

' Cape Observations, p. 889. ' AsUr, Jour. vol. ii. p. 101. 

■ MonthJ/y Notices, vol. zxz. p. 50. 
See Mr* J. B. Sutton'a remarks, lUustirated Science MofUhly, voL U. 
pp.68, 199* 


in the Milky Way. If these were due to the interlacing by 
perspective of branches really far apart in space, the enclosing 
nebula should be markedly fainter on one side than on the 
other. But this is not so. The borders of the southern Coal- 
sack, for instance, are approximately of the same brightness 
all round. A single vivid mass has obviously been the scene 
of what, in the absence of better knowledge, may be described 
as an excavatory process. 

Again, on the spiral theory, the great rift in the Milky 
Way should be ordinary sky-background, the branches on 
either hand being mutually disconnected except through the 
optical effect of projection. But there can be no doubt that 
the rift forms, in a certain sense, a part of the galactic struc- 
ture. Stars are much more thickly strewn upon it than in 
the external heavens. Argelander, in fact, showed that, down 
to 9*5 magnitude, they are only -^ less plentiful than in the 
adjacent branches of the Milky Way itself, and are actually 
more plentiful than in the section of the undivided stream 
passing from Perseus to Auriga.^ Nor can the fissured parts be 
regarded as truly independent. Their separation is gradually 
effected. Premonitory cavities seem to announce it before- 
hand, and even after it has become definitive, abortive efforts 
towards reunion are indicated by the correspondence of oppo- 
site projections. The bifurcation is beyond question a physical 

Over-subtlety has been the besetting snare of theorists on 
this important subject. Perhaps by adopting the simple view 
that the Milky Way is very much what it seems, we shall get 
nearer to the truth than by indulging in more recondite 
speculations. What it seems to be — especially as delineated by 
Dr. Boeddicker — is a ring with streaming appendages — out- 
liers extending from the main body in all possible directions, 
some nearly straight towards, or away from us, others at 
every imaginable angle with our line of sight. The results in 
perspective foreshortening must evidently, under these circum- 
stances, be highly complex; the eye being presented with 
groups and streams of stars, at immensely different real dis- 

^ B<m/ner.Meob,B^Y.SiinkUtmg. 


iances, but all projected indiscriminately upon the same zone 
of the heavens. Thus, while some branches, pursued along 
their outward course, fade at last into dim nebulosity, other 
Milky Way groups may be distinguished as bright separate 
stars, because much nearer to us than the generaUty of their 

The internal organisation of the Galaxy must be in the 
last degree intricate. It collects within its ample round, there 
is every reason to suppose, an absolutely endless variety of 
separate systems. A multitudinous aggregate of individual 
clusters, it exhibits, moreover, as a whole, the structure of one 
single cluster on a prodigious scale. Its fringed edges, its 
rifts and vacuities, are, as we have seen, reproduced in 
miniature in innumerable star-groups. ' Bings,' and * sprays,' 
and * streams ' of stars are unmistakably common to the two 
orders of formation ; and the stellar constituents of both are 
frequently involved with gaseous nebulsB in a way showing 
most intimate association by origin and development. The 
laws then governing stellar aggregation in the one case govern 
it also in the other ; and so, from this direction independently, 
we again reach Herschel's conclusion that the Milky Way 
* consists of stars very differently scattered from those which 
are immediately about us.' 

But are these stars sum, co-ordinate with our own ? or 
must we regard them as comparatively insignificant bodies, 
sharing a sun-like nature, indeed, but on a far lower level of 
power and splendour? The question is equivalent to this 
other perpetually recurring one, What is their average distance 
from ourselves ? In what portion of space do the true galactic 
condensations occur? How far outward should we have 
to travel before finding ourselves actually in the midst of 
the crowded objects producing, to terrestrial observers, the 
' milky ' effect of a nebulous stratum ? 

Now we are not so completely destitute of knowledge on 
this point as is commonly supposed. Our readers, we believe, 
will, when the nature of the evidence already at hand has 
been explained to them, readily admit its all but demonstra- 
tive significance to the effect that the star-clouds of the 

THE miLky way $66 

galaxy lie beyond the average distance of tenth-magnitude 

The numbers of each order of stars down to the ninth 
magnitude, derived with approximate precision from Arge- 
lander's ' Durchmusterung/ are found, on the whole (apart 
from an excess of bright stars to be separately accounted for), 
to increase at about the rate to be expected on the hypothesis 
of an equable distribution through space of bodies on the 
same general level of luminous power. This kind of uniformity 
does not of course preclude any extent of individual variety ; 
it only means that individual varieties, however wide, balance 
each other in the long 'run, and on a sufficiently extended 
average. The stars then, down to the ninth magnitude, are 
distributed roughly according to the space they occupy ; there 
are more of each magnitude in proportion to the increased 
cubical contents of the successive spheres, the radii of which 
are the theoretical mean distances corresponding to each suc- 
cessive magnitude. This theoretical increase of numbers con- 
currently with available space, is measured by the cube of the 
distance-ratio; it is nearly four-fold. That is to say, there 
should be (supposing equable distribution) close upon four 
times as many stars of a given magnitude as of the magni- 
tude immediately superior to it. And, speaking roughly, 
there are. This law of increase, moreover, seems to be con- 
formed to, as closely as could be expected, over the galactic 
zone itself no less than over the rest of the sky. 

It is then perfectly evident that, down at least to the ninth 
magnitude, the progression outward is unbroken by any great 
or systematic accumulation of stars. The Milky Way stcarms 
collect further off. But there is more. The stars of the 
' Durchmusterung,' including a multitude of nominally 9'5, 
but really tenth magnitude, are, as already mentioned, nearly 
as thickly strewn over the dark fissure of the Milky Way 
between Cygnus and Centaur, as over the bright enclosing 
branches. The nebulous effect of these latter to the eye is 
then presumably due to more remote collections. As to the 
further limits of these, we know as yet nothing, except that 
Herschel's gauge-numbers left it to be inferred that 'thinning 




The question whether nebulae are external galaxies hardly 
any longer needs discussion. It has been answered by the 
progress of discovery. No competent thinker, with the whole 
of the available evidence before him, can now, it is safe 
to say, maintain any single nebula to be a star system of co- 
ordinate rank with the Milky Way. A practical certainty has 
been attained that the entire contents, stellar and nebular, 
of the sphere belong to one mighty aggregation, and stand in 
ordered mutual relations within the Umits of one all-embracing 
scheme — all-embracing, that is to say, so far as our capacities 
of knowledge extend. With the infinite possibilities beyond, 
science has no concern. 

The chief reasons justifying the assertion that the status 
of the nebulae is intra-galactic, are of three kinds. They 
depend, first, upon the nature of the bodies themselves; 
secondly, upon the stellar associations of many of them ; 
thirdly, upon their systematic arrangement as compared with 
the systematic arrangement of the stars. 

The detection of gaseous nebulae not only directly demon- 
strated the non-stellar nature of a large number of these 
objects, but afforded a rational presumption that the others, 
however composed, were on a commensurate scale of size, and 
situated at commensurable distances. It may indeed turn 
out that gaseous and non-gaseous nebulae form an unbroken 
series, rather than two distinct classes separated by an im- 
passable barrier. For the bright lines indicating gaseity are 
accompanied by more or less of continuous light, and the 
continuous spectra significant of advanced condensation 
perhaps include bright lines, while the proportions of these 


continuous and discontinuous ingredients differ considerably 
in different individual nebulae. But before any settled opinion 
can be formed as to whether these differences have really the 
transitional or * evolutionary ' meaning we might be inclined 
to attribute to them, nebular spectroscopy must be a good deal 
further advanced than it is at present. Apart from this 
question, however, there is such strong evidence of relation- 
ship between the various orders of nebulae that to admit some 
to membership of the sidereal system while excluding others 
would be a palpable absurdity. And since those of a gaseous 
constitution must be so admitted, the rest follow as a matter 
of course. 

Of the physical connection of nebulae with particular stars, 
fresh and incontrovertible proofs accumulate day by day. 
Nothing can be more certain than that objects of each kind co- 
exist in the same parts of space, and are bound together by 
most intimate mutual ties. To argue the matter seems, as the 
French say, like ' battering in an open door.' We need only 
recall the stars of the Pleiades, photographically shown to be 
intermixed with nebulae, and those in Orion still bearing in 
their spectra traces of their recent origin from the curdling 
masses around. The nuclear positions so frequently occupied 
in nebulae by stars single and multiple, reiterate the same 
assertion of kinship, emphasised still further by the phenomena 
of stellar outbursts in nebulae. The scepes of these vimt lie 
within the circuit of the Milky Way, unless we are prepared 
to assume the occurrence, in extra-sidereal space, of confla- 
grations on a scale outraging all probability.* It has been 
calculated that if the Andromeda nebula were a universe apart 
of the same real extent as the Galaxy, it should be situated, 
in order to reduce it to its present apparent dimensions, at a 
minimum distance of twenty-five galactic diameters.^ And a 
galactic diameter being estimated by the same authority at 
thirteen thousand light-years, it follows that, on the supposi- 
tion in question, light would require 325,000 years to reach 
us from the nebula. The star then which suddenly shone out 

' This point was frequently insisted upon by the late Mr. Proctor. 
« Weisse, Schriftm Wiener VereinSj Bd. v. p. 318. 

B B 


in the midst of it in August 1885, should have been at 564 
times the distance inferrible from its effective brightness. In 
real light it should have been equivalent to 818,000 stars like 
Regulus, or to nesirlyjifty million such suns as our own ! But 
even this extravagant result inadequately represents the real 
improbability of the hypothesis it depends upon ; since the 
Andromeda nebula, if an external galaxy, would almost cer- 
tainly be at a far greater remoteness from a sister-galaxy 
than would be represented by twenty-five of its own diameters. 

Just as the Milky Way might be described as a great com- 
pound cluster made up of innumerable subordinate clusters, 
so the greater Magellanic Cloud seems to be a gigantic nebula 
embracing, and bringing into some kind of correlation, multi- 
tudes of separate nebulae. To the naked eye it shows vaguely 
a brighter axis spreading at the extremities so as to produce a 
resemblance to the * Dumb-bell ' nebula ; it shows, that is to 
say, signs of definite organisation as a united whole ; and it 
includes, strangely enough, among its inmates a miniature of 
itself (N. G. C, 1978), but of much greater intensity and dis- 
tinctness. Sir John Herschel's enumeration in 1847 of the con- 
tents of the * Cloud ' gave conclusive evidence of the interstellar 
situation of nebulsB — evidence the full import of which Dr. 
Whewell was the first to perceive. Over an area of forty-two 
square degrees, 278 nebular objects (stars being copiously inter- 
spersed) are distributed with the elsewhere unparalleled density 
of 6^ to the square degi-ee. * The Nubecula Major,' Herschel 
wrote, * like the Minor, consists partly of large tracts and ill- 
defined patches of irresolvable nebula and of nebulosity in every 
stage of resolution, up to perfectly resolved stars like the Milky 
Way, as also of regular and irregular nebulae properly so 
called, of globular clusters in every stage of resolvabihty, and 
of clustering groups sufficiently insulated and condensed to 
come under the designation of clusters of stars.' ^ 

Here then we find— in a system certainly, as Herschel said, 
* 8ui generis,' yet none the less, on that account, instructive as 
to cosmical relationships — undoubted stars and undoubted 
nebulae at the same general distance from the earth. Some 

> Cape Rcsidts, p. 146. 


of the nebulfiB may indeed very well be placed actually nearer 
to us than some of the stars ; and the extreme possible differ- 
ence of their remoteness cannot in any case exceed one-tenth 
of the interval between the hither edge of the Cloud and 
ourselves. We learn too the plain lesson that distance is only 
one factor in the production of ' irresolvabihty.' For stars in 
every stage of crowding, from loose groups to the veriest dust- 
streaks, globular clusters coarse and fine, nebulas of all kinds 
and species, range side by side in this extraordinary collection, 
proving beyond question that differences of aggregation are 
real and enormous, and need no additional abysses of space 
to account for them. 

Even, however, if all these mutually confirmatory argu- 
ments could be dismissed as invalid, the mode of scattering 
of nebulflB on the sky-surface would alone suflSce to demon- 
strate their association with the sidereal system. Sir William 
Herschel was early struck with the occurrence of beds of these 
objects, preceded and followed by spaces void of stars. His 
assistant was indeed sometimes warned by him, not without 
good cause, * to prepare, since he expected in a few minutes 
to come at a stratum of the nebulae, finding himself already 
on nebulous ground.' * He attained, too, a partial comprehen- 
sion of the larger plan of their distribution, as being the inverse 
to that of stars ; but the younger Herschel first brought into 
clear view the distinct and striking division of the nebulae 
into ' two chief strata, separated by the Galaxy.' Taking the 
circle of the Milky Way as a horizon, he remarked that the 
mass of them gathered together in Virgo and Coma Berenices 
'forms, as it were, a canopy occupying the zenith, and 
descending thence to a considerable distance on all sides, but 
chiefly on that towards which the (celestial) north pole lies.' * 

This characteristic of accumulation about the galactic pole 
is less marked in the southern hemisphere, though here too 
there is a * chief nebular region ' approximately corresponding 
to that in Virgo. The distribution is, however, on the whole, 
much more uniform than in the northern hemisphere, or 
rather, more uniformly patckyy rich districts alternating with 
* PhU. Trans, vol. Ixxiv. p. 449. " Caye BestUts, p. 137. 

B B 2 


more or less ample vacuities. One of these extends about 
fifteen degrees all round the south pole, the Lesser Cloud 
marking its edge. The remarkable fact, too, was noticed 
by Sir John Herschel that the larger nubecula seems ^ to 
terminate something approaching to a zone of connected 
patches of nebulaB,' reaching across Dorado, Eridanus, and 
Cetus to the equator, where it unites with the ' nebular region 
of Pisces.' A similar line of communication is less con- 
spicuously kept open with the minor nubecula, and this 
feature of * streams ' of nebulsB with terminal aggregations was 
considered by Mr. Proctor to be distinctive of southern skies.* 
He adverted besides to the coincidence of two of them with 
stellar ' streams ' in Eridanus and Aquarius, and was struck 
with a significant deficiency of bright stars over the intervals 
between nebular groups." 

The general facts of nebular distribution were correctly 
described by Mr. Herbert Spencer in 1864. ^ In that zone,' 
he wrote, * of celestial space where stars are excessively 
abundant, nebulae are rare ; while in the two opposite 
celestial spaces that are furthest removed from this zone 
nebulae are abundant. Scarcely any nebulae lie near the 
galactic circle; and the great mass of them lie round the 
galactic poles. Can this be mere coincidence ? When to the 
fact that the general mass of nebulae are antithetical in posi- 
tion to the general mass of stars, we add the fact that local 
regions of nebulae are regions where stars are scarce, and the 
further fact that single nebulae are habitually found in com- 
paratively starless spots, does not the proof of a physical 
connection become overwhelming ? ' ^ 

Accompanying, but considerably overlapping the Milky 
Way along its entire round, is a ' zone of nebular dispersion ' 
(as Mr. Proctor called it) — a wide track of denudation, so far 
as these objects are concerned. The nebular multitude shrinks, 
as it were, from association with the congregated galactic 
stars. A relation of avoidance is strongly accentuated. But 
withdrawal implies recognition. It implies the subordination 

' M<mthly Notices, vol. xxix. p. 340. « Ibid. p. 344. 

• The Nebular Hypothesis (with Addenda), p. 112. 


of stars and nebulaB alike to a single idea embodied in a single 
scheme. Of our possible acquaintance, then, there is but one 
* island universe ' — that within whose boundaries our temporal 
lot is cast, and from whose shores we gaze wistfully into in- 

Dismissing, then, the grandiose but misleading notion that 
nebulsB are systems of equal rank with the Galaxy, we may 
turn our attention to the problems presented by their situation 
within it. When the facts connected with it are looked at in 
detail, distinctions become evident between the different classes 
of nebulae— distinctions so marked as to lead almost to their 

The ' relation of avoidance ' to the Milky Way just adverted 
to prevails only among the ' unresolved ' nebulaB. These, it is 
true, are the great majority of the entire, so that the conclu- 
sion of nebular crowding away from that zone remains un- 
impeachable. For certain classes of minor numerical, but 
high cosmical importance, the relation is precisely inverted. 
Over gaseous nebulae and clusters, the Milky Way seems to 
exercise an attractive influence equally strong with its repulsive 
effect upon nebulae of all other kinds. 

Forty out of one hundred and two globular clusters belong 
to the galactic zone,^ which is hence thirteen times more richly 
furnished than the rest of the sky with this peculiar descrip- 
tion of objects. And the excess rises to forty-two times for 
irregular or nondescript clusters, 434 out of 535 of which — 
that is, eighty-one per cent. — are located in, or close to, the 
Milky Way. Many clusters, indeed, obviously form an integral 
part of the formation itself; of others, it is diflScult to decide 
whether they should be ranked as distinct, or simply as in- 
tensifications of ordinary galactic star- groupings. To the 
latter category almost certainly belongs a collection (M 24) 
visible to the naked eye as a dim cloudlet near fi Sagittarii, 
and named by Father Secchi, *Delle Caustiche,' from the 

* Taken as of the nniform width of thirty degrees, and covering i of the 

sphere. Major Markwick (Jour, Liv, Asir. Soc. vol. vii. p. 182) finds the pro- 
portionate area of the Milky Way in the northern hemisphere to be J, , in the 
sonthem ^. 


peculiar arrangement of its stars in rays, arches, caustic 
curves, and intertwined spirals. This again is included in 
the great oval condensation of galactic stars, shown (from a 
Lick photograph) in fig. 42, and obviously endowed with some 
degree of structural independence. 

Gaseous nebulae, like gaseous stars, are nearly exclusive in 
their galactic affinities.* Very few planetaries can be found 
at any considerable distance from the favoured zone; the 
spectroscopic search for stellar nebulae is fruitless unless within 
its borders ; and they embrace — with one exception — aU the 
irregular nebulae. This single exception is a most significant 
one. It is that of the 'great looped nebula,' one of the 
numerous constituents of the greater Magellanic Cloud. 
Plainly, then, the same peculiar conditions which have allowed 
primitive cosmical matter to remain imcondensed in galactic 
regions prevails also in the nubecula, although here they are 
found consistent with the presence, in large numbeis, of those 
species of nebulae in great measure banished from the Milky 

Within its precincts only one in sixteen of those dim, often 
fantastically-shaped, objects is met with, the analysed light of 
which gives no indication of gaseity, while their even texture, 
under the highest telescopic powers, suggests no approach to 
the stage of breaking up into stars. What then is their 
nature? Is the difference separating them in appearance 
from the resolvable nebulae, i.e. clusters, crowding the Milky 
Way, a diflference of distance solely ? Are they, too, clusters of 
a further degree of remoteness, and therefore inaccessible to 
effective scrutiny ? There is nothing in their aspect to pre- 
clude this supposition. So far as observation can tell, they 
may be of stellar composition. Only it is not easy to under- 
stand why the average distance of nebulae situated near the 
galactic poles should be many times that of nebulae thronging 
the vicinity of the galactic equator. 

Mr. Cleveland Abbe ^ imagines the nebulae to be equably 
distributed over the surface of a * prolate ellipsoid,' its longer 

* Bauschinger, V. J, S. Astr, Ges. Jahrg. xxiv. p. 43. 

* Monthly NoUcea, voL xxvii. p. 262, 


axis coinciding approximately with the axis of the Milky Way ; 
and this arrangement would undoubtedly give an appearance 
of crowding in the observed directions, since to an eye placed 
near the centre of such an oval figure, objects uniformly 
scattered over its surface would produce, by perspective, the 
effect of running together near its pointed ends. But it is 
scarcely credible that things should in reality be disposed in 
this highly artificial manner. It is not enough barely to * cover 
the phenomena ' with a theory, unless the theory be in itself 
congruous with the general plan of operations upon which we 
can see that nature works. Besides, the local distribution of 
nebulflB is so far from uniform, that antecedent probability is 
in favour of their general distribution, too, being marked by 
striking irregularities. The ' canopy ' of nebulae in Virgo is, 
then, we may rest assured, as genuine an accumulation in its 
own way as the spherical assemblage in the Magellanic Cloud. 
But if there be no systematic difference of distance between 
the nebular classes occupying contrasted situations as regards 
the lines of galactic structure, there must be a systematic dif- 
ference of constitution.* The parts of those objects crowding 
towards the poles must be comparatively small and close 
together. We have indeed already found reason to believe 
that clusters do, in point of fact, merge insensibly into nebulas 
— that groups of genuine suns at wide intervals stand at the 
summit of an unbroken gradation of systems with continually 
smaller and closer constituents, down to accumulations of 
what is almost literally * star-dust.' Eesolvability is then, we 
repeat, a question of constitution quite as much as of distance, 
and we are brought to the conclusion that, while galactic 
nebulsB are of what we may roughly describe as stellar com- 
position, non-galactic nebulae are more or less pulverulent. 
We cannot of course pretend to account for this remarkable 
distinction. All that can be said is that it appears to be 
actually existent. The irresolvable * polar ' nebulae perhaps 
escaped influences powerful over the * equatorial ' ones. Their 
development, at any rate, seems to have taken a differen 

* Proctor, Monthly Notices, toI. xzii. p. 342 


No direct proof of motion in nebulae has — as we have seen 
in an earlier chapter * — yet been obtained. In no single case 
has either visual change of place or spectroscopic alteration 
due to recession or approach been ascertained. But the mode 
of nebular distribution affords indirect evidence of movement. 
* Streaminess/ if it mean anything, implies that the bodies 
affected by it advance in common towards a common goal. 
Aggregation at the end of a stream prompts the conjecture 
that a motion of advance was at a certain point, by some 
supervening attraction, swayed into a motion of revolution. 
A hint as to the origin of the Magellanic Clouds may hence 
be derived. They represent in some sort vessels filled through 
long pipes from a vast reservoir. And since the pipes are 
still there, the flow may be conceived to be still in progress. 
Were it to cease, the connection of the nubeculee with the 
main nebular body would eventually be interrupted, and their 
insulation would become complete. 

The fidelity with which gaseous nebulae and clusters 
adhere to the Milky Way as seen projected upon the sphere 
warrants the inference that their distribution in space is of a 
similar character. It would be unreasonable to disconnect 
them from a formation of which they so closely follow the 
lines. We can scarcely err in supposing that they lie in 
general within, not behind or in front of it. Thus, the 
globular clusters richly strewn over the branch of the Milky 
Way from Scorpio to Ophiuchus, but withdrawn from the 
conterminous dark rift, plainly belong to the aggregations 
of minute stars, owing to the absence of which the fissure 
seems black, although copiously stocked with stars to the 
tenth magnitude. Other condensed groups stand out from a 
curtain of apparently still more remote stars, representing 
possibly a divergent galactic ramification. 

Such ramifications must in many cases be greatly fore- 
shortened as viewed from our nearly central position; in 
some, they may appear only as brilliant knots upon the 
* trunk ' of the Milky Way. Possibly, the double cluster in 
Perseus may partake of this optical character. It may be 
* See antct p. 290. 


the termination of a branch spreading inward, and seen 
nearly end-on. Its constituent stars ajypear to be at vastly 
different distances from our eyes. The appearance perhaps 
corresponds with the reality. But this is a mere conjecture, 
and one which there is no immediate prospect of testing by 
comparison with facts. 

Like the Perseus clusters, the Orion nebula gives in- 
dications of greater proximity than the main galactic accumu- 
lation, to which it is nevertheless beyond doubt structurally 
related. For a winding nebulous extension from the Milky 
Way can be traced past a Orionis in the right shoulder through 
the belt and sword, the bright stars marking which are de- 
monstrably associated with the nebula. The inference then 
presents itself that the whole mixed system, or series of 
systems, is placed upon an obliquely directed offset from the 
galactic zone. Keasoning of the same kind may perhaps 
apply to the combined nebula and cluster M 8. It occurs 
as a premonitory outlier of the leading division of the 
fissured Milky Way, from which it lies a Kttle apart ; and it 
seemed to Sir John Herschel only an ' intense exaggeration ' 
of the stellar collections in its neighbourhood.* 

Summarising our conclusions, we find the unity of the 
stellar and nebular systems to be fully ascertained. They 
are bound together by relations of agreement and contrast 
scarcely less visibly intimate than those severally connecting 
individual members of each order. The general plan of 
nebular distribution is into two vast assemblages, one on 
either side of the galactic zone; but while this is, com- 
paratively speaking, avoided by the unresolved crowd, it is 
densely thronged with clusters and gaseous nebulse. The 
conditions of aggregation within the zone are hence inferred 
to differ from those prevailing outside it ; but in what respect 
they are different cannot readily be surmised, far less deter- 
mined. As to the distances of the nebulae, we know nothing 
positive; they no doubt vary extensively; nor can either 
fineness of grain or faintness of light (both of which may be 
inherent qualities) serve to distinguish between those nearest 

> Cape Results, p. 887, 


to, and those further away from us. We may, however, 
plausibly conjecture that the hood-like accumulations of the' 
nebulae are of about the same order of remoteness with 
eleventh or twelfth magnitude stars, thus constituting, as it 
were, polar caps on a sphere of which the annular formation 
of the Milky Way marks the equator. 




Sir William Herschel conceived it to be the supreme object of 
astronomy * to obtain a knowledge of the construction of the 
heavens;' and this, in his view, would be accomplished by 
the * determination of the real place of every celestial body in 
space.' * Thus limited, the problem would be completely 
solved could the absolute distance be ascertained of every 
object telescopically or photographically discernible in the sky. 
But even the attainment of this unattainable point would never 
have satisfied Herschel's restless spirit. The real scope of his 
inquiries went far beyond it. They had an historical, as well 
as a statistical aim. * Looking before and after,' they embraced 
the past and future, no less than the present of the Cosmos. 

Modern investigators are of the same mind. The heavens 
are regarded by them from a physiological, rather than from a 
purely anatomical point of view. Mere knowledge of structure, 
however accurate, will not content them. The vital functions 
of the organism, the mutual dependence of its parts, the 
balance of internal forces tending towards destruction and 
preservation, the dimly-apprehended aim of its divinely sus- 
tained activity, engage their eager attention. The heavens 
live and move, and the laws of their life and motion involve 
the material destiny of man. It is impossible that he should 
be indifferent to them. 

Even, however, if our instinctive interest in the working 
of the machine were less keen, we should be driven to search . 
out the dynamical relations of its parts by the impossibility of 
otherwise arriving at a true knowledge of their geometrical 
relations. Not only are these variable from one moment to 

> PkU. Trans, vol. cviL p. 302. 


another, but acquaintance with them at any single moment is 
not conceivably accessible to us apart from previous acquain- 
tance with modes and laws of motion. For our view of 
sidereal objects is not simultaneous. Communication with 
them by means of light takes time, and postdates the sensible 
impressions by which we are informed of their whereabouts, 
in the direct proportion of their distances. We see the 
stars not where they are — not even where they were, at any 
one instant, but where they were on a sliding scale of instants. 
The epoch corresponding to the apparent position of each is 
different, and the range of difference extends over many 
thousands of years. The reduction of those positions to a 
common epoch so as to get a survey of the genuinely con- 
temporary relations in space of all sidereal objects — ideally 
feasible at best — could not so much as be thought of as 
possible without a preliminary knowledge of their displace- 
ments during the centuries, or millenniums, elapsed since the 
ethereal vibrations they originate started on their several 
journeys hither. Thus the study of configurations blends 
with the study of movements and forces; the restrictions 
placed upon thought by the effort to exclude all but a single 
aspect of phenomena fall away of themselves, and we are 
confronted, whether we will or no, by the stupendous problem 
of the universe as a vital whole. 

As a whole; but not necessarily as the whole. The 
sidereal world presents us, to all appearance, with a finite 
system. Human reason would, indeed, otherwise be totally 
incompetent to deal with the subject of its organisation. There 
would be nothing for it but to lay down the arms of our 
understanding before its transcendental and appalling magni- 
tude. But the probability amounts almost to certainty that 
star-strewn space is of measurable dimensions. For from 
innumerable stars a limitless sum-total of radiations should 
be derived, by which darkness would be banished from our 
skies ; and the * intense inane,' glowing with the mingled 
beams of suns individually indistinguishable, would bewilder 
our feeble senses with its monotonous splendour. This laying 
bare, so to speak, of the empyrean would be the simple and 


certain result of the continuance ad infinittim of any arrange- 
ment of sidereal objects comparable with that prevailing in 
our neighbourhood. Unless, that is to say, light suffer some 
degree of enfeeblement in space. If this be the case, then 
our reasonings are put to silence, and a veil is drawn im- 
penetrable to scrutiny. But there is not a particle of evidence 
that any such toll is exacted ; contrary indications are strong ; 
and the assertion that its payment is inevitable depends upon 
analogies which may be wholly visionary.* We are then, for 
the present, entitled to disregard the problematical effect of a 
more than dubious cause. The sidereal system cannot accord- 
ingly be regarded as in any true sense infinite. The scale 
upon which it is constructed baffles, it is true, the utmost 
strain of the imagination to conceive ; in the multitudinous 
splendour of its components, in the number and variety of 
the subordinate groups constituted by them, in the magni- 
ficent play of forces it unfolds, in the dim processes of develop- 
ment it suggests, it bears glorious witness to the power and 
wisdom of the Almighty Designer ; yet it has limits, and for 
that reason it is a fit subject for the exercise of limited 
understandings. With further systems, * pinnacled deep ' out 
of our sight for ever, we have, properly speaking, no scientific 
concern ; we only know that * when a man hath done, then 
shall he begin ' to declare the wonderful works of God. 

Eegarding the visible world of stars and nebulae as an 
isolated, though excessively complex, system, we may try to 
give the best order we can to our ideas respecting its consti- 
tution. Let us see what are the available data. The number 
of stars actually registered, as stated in the opening chapter 
of this work, is about 650,000,'^ of which three-quarters, or 
thereabouts, are of magnitudes between the ninth and tenth, 
and the rest are brighter. Beyond the Umits of this great 
census, minute stars aboimd ; but to how many millions they 
would sum up if completely enumerated, cannot even be 
guessed with a show of probability. Sir John Herschel esti- 

* Hirn, Constitution de VEspace Cdeste, p. 297. 

'^ Inclnding those in Dr. Gill's southern ^Durchmusterung,' the photo- 
graphic work for which was completed in 1889, while the resulting catalogue 
may be looked for in a couple of years. 


mated at five and a half millions the stars (to the fourteenth 
photometric magnitude) perceptible over the entire sky with 
his twenty-foot reflector; Struve calculated them at twenty 
millions ; and it has been vaguely surmised that a hun- 
dred millions could be shown by the most powerful modern 
telescopes. The truth is, that we are still almost wholly 
ignorant on the point. Different parts of the sky vary 
enormously in richness. In some, telescopic stars literally 
swarm ; in others, they occur by comparison scantily. It has 
been computed by Mr. Gore * that if the whole heavens were 
as thickly strewn as the region of the Pleiades, the number of 
stars to the seventeenth (nominal) magnitude, would be about 
thirty-three millions. But the method of distribution within 
a definite cluster evidently gives no clue to that prevaiUng 
outside it. A fair specimen-field is, indeed, all but impossible 
to choose. Counts in the Milky Way, extended in the same 
proportion over the sphere, would enormously exaggerate the 
crowding of the stars ; which would, in an equal degree, be 
underrated by counts executed apart from it. 

Eeliable data on the subject can only, it would seem, be 
collected with practical usefulness by the method of * photo- 
graphic star-gauging.' Reckonings of the stars in their 
light-ranks, upon plates exposed for various lengths of time, 
ought to tell with certainty how far the ideal law of augment- 
ing numbers holds good, and where * thinning-out ' becomes 
apparent. In an equable stratum (as our readers are aware),* 
the stars must nearly quadruple at each descent of a mag- 
nitude, simply because the cubical space holding them is 
quadrupled. Should this rule be overthrown by excess, a real 
aggregation is indicated, at the distance corresponding to the 
altered rate of increase ; if by defect, then obviously the supply 
of stars in the region examined is becoming exhausted, their 
scattering is sparser than in our nearer vicinity, and the ter- 
mination of the series is at hand, if not already reached. 

Now even HerscheFs Milky Way gauges afforded such in- 
distinct evidence of a terminating series as alone could be de- 
rived from aggregate counts, in the fact that the numbers of 

> Jour, Liv, Astr, Soc, vol. vii. p. 180. * See ante, p. 365. 


stars recorded by him amounted to only one-third of what 
might have been anticipated from the penetrating power of his 
instrument applied to an indefinitely extended system. And 
for a ' mean . sounding/ at the northern galactic pole, M. 
Celoria, with a refractor showing, at the utmost, eleventh - 
magnitude stars, obtained a number almost identical with that 
given byHerschel's great reflector. The larger instrument, 
then, here revealed no additional stars. Similar symptoms of 


ICftgs. 9-0 9-5 10-0 10-6 11-0 11-6 13-0 13'6 13-0 13-5 UD U'S 16-0 

Fio. 50.— Distribntion of 984 stars within 1^ of the pole, showing the ratio 
of numbers to space for each half -magnitude. 

exhaustion in the star-supplies may be found in Professor 
Pickering's photographic catalogue of 947 stars within one 
degree of the celestial north pole.^ A single glance at the 
synoptical table giving the numbers for each half-magnitude 
suflBces to show that the numerical representation of the lower 
ranks is inadequate. The small stars are overwhelmingly too 
few for the space they must occupy if of average brightness ; 

* Harvard Annals, vol. xviii. p. 202. 


and they are too few in a constantly increasing ratio. The 
accompanying diagram (fig. 50) represents graphically the de- 
crease outward of density (or the proportion of numbers to 
space), deducible from Professor Pickering's enumeration on 
the sole supposition of the equal average lustre of each class 
of stars. Those of the ninth are the most thickly strewn ; 
the intervals between star and star widen rapidly and con- 
tinuously (for the sudden dip at 9*5 magnitude is evidently 
accidental) down to 11-5 magnitude, when a slight recovery, 
lasting to the thirteenth magnitude, sets in. To what extent 
these fluctuations are of a systematic character, can only be 
decided by more comprehensive surveys ; for the present they 
serve to make it clear that in some parts of the sky faint stars 
fall far below, while in others they perhaps largely exceed 
their due proportions. 

The influence of the Milky Way is predominant over the 
general distribution of the stars, but it grows more marked 
with their diminishing brightness. Its inferior efficacy in the 
southern hemisphere ^ may perhaps be regarded as correspond- 
ing to the lower degree of polar condensation visible in the 
southern, as compared with the northern nebulsB. The pre- 
ference of lucid stars for the Milky W&y is so slight that it 
might easily be overlooked ; but it appears from some careful 
statistics published by Mr. Gore* that even in them the 
galactic zone is one and a half times richer than other parts 
of the sky. There is some evidence, however, that this crowd- 
ing is towards a plane of condensation distinct from, though 
very close to, that of the galaxy. 

A zone of large stars traversing the southern hemisphere 
was thought by Sir John Herschel to be the projection of * a 
subordinate sheet, or stratum, deviating some twenty degrees 
from parallelism to the Milky Way.' ^ The hint was further 
developed by Dr. Gould. * Few celestial phenomena,' he con- 
sidered to be * more palpable than the existence of a stream or 
belt of bright stars,' traceable ' with tolerable distinctness 
through the entire circuit of the heavens, and forming a great 

» Seeliger, Sitgungsberichte, Munich, 1884, p. 521, 1886, p. 220. 

« Jot4/r, Liv. Astr. 8oc. vol. vii. pp. 175, 182. ' Gape Results, p. 385. 


circle as well defined as that of the galaxy itself,' * which it 
crosses at an angle of about 20° in Crux and Cassiopeia. 
Traversing in the southern hemisphere Orion, Canis Major, 
Argo, the Centaur, Lupus, and Scorpio, it pursues its way in 
the northern through Taurus, Perseus, Cassiopeia, Cepheus, 
Cygnus, and Lyra, its line being less obviously continued by 
the stars of Hercules and Ophiuchus.* Like the Milky Way, 
it seems to bifurcate near a Centauri, the branch there thrown 
off reuniting with the parent stem in Andromeda. That the 
stars thus marked out, to the number of about five hundred, 
constitute with the sun a cluster * of a flattened and somewhat 
bifid form,' ' distinct from the vast organisation of the Milky 
Way,' grew into a conviction with the progress of Dr. Gould's 
observations. Since, moreover, they were found, on fuller 
inquiry, to trace on the sphere only a small circle, to the 
south of the parallel great circle, the sun's place within the 
group must be removed to the north of its medial plane, 
towards Leo, and is supposed to be nearly marginal in the 
direction of Hercules and Ophiuchus, where the track of 
the 'belt' is almost effaced by the wide scattering of the 
constituent stars. 

Two circumstances appear to show that this * solar 
cluster ' is in some way a reality. In the first place, an 
enumeration of the stars in photometric order discloses a 
systematic excess of objects brighter than the fourth magni- 
tude, making it certain that there is actual condensation in 
the neighbourhood of the sun — that the average allowance of 
cubical space per star is smaller within a sphere enclosing 
him with a radius, say, of 140 light-years, than further 
away. Again, when deduction has been made of some five 
hundred stars of the higher orders, the remainder, to the 
ninth magnitude, form a tolerably regular series, increasing 
in numbers nearly in the theoretical proportion of their 
diminishing light.^ And the conclusion that these five 

' Amer. Jour, of Science^ vol. viii. p. 333 (1874). ' XJr(m.ArgenHnay p. 365. 

* The * empirical ratio * (that resulting from actual enumeration) of multi- 
plication of numbers per magnitude, is 3*912 ; the theoretical ratio is 3'987. 
— XJran, Argentina^ p. 367. 



hundred superfluous objects do in fact compose a group 
apart, is strengthened by the symmetrical arrangement, 
with regard to the 'belt/ of the bright stars outside it. 
Their tendency to collect towards its central plane is, accord- 
ing to Dr. Gould, irrespective of the Milky Way, except in so 
far as the two formations coincide by the projection of one 
upon the other. 

It is worth notice, too, that the present direction of the 
solar movement agrees with the general ' lie ' of the belt ; 
although the other stars it is supposed to include give no sign 
of conformity to any fundamental plane of revolution. The 
truth of the hypothesis can, however, be tested only by a 
detailed study of motions, presumably of extreme complexity. 
The problem indeed of separating the individual displacements 
of the clustered stars from those common to all and shared 
by ourselves, as well as from the independent shiftings of 
extraneous objects, baffles all the present resources of science. 
The ascertained translation of the sun must, if the * solar 
cluster' be really organised on a separate dynamical basis, 
be mainly ' interstitial ' ; it must represent the revolution of 
the solar system round the centre of gravity of the col- 
lection ; for it emerged, essentially as we know it now, from 
Herschel's interpretation of the apparent displacements of 
a few undoubted members of the group. The application of 
the same method (as recommended by Professor Holden)* to 
stars unequivocally external to the group, would, if practicable, 
be of extreme interest. Thus, if at all, the bodily transport 
through space of the entire solar cluster might be determined, 
and Herschel's forecast realised of a still higher grade in the 
hierarchy of motions than that corresponding to the sun's 
journey towards X Herculis. It is even conceivable that such 
a movement of the entire stellar system of the sun with refer- 
ence to other similar systems, or to what Herschel called ' inter- 
systematical stars,' might account for an otherwise inexpU- 
cable excess of movement among the smaller stars,^ the apparent 
displacements of which would be increased by a perspective 
element absent from those of brighter and nearer objects. 

' Century Magazine^ Sept. 1889. * See ante, p. 338. 


The attempt, however, ideally to compoBe the solar cluster 
encounters enormous difficulties. Should, for instance, the 
stars of Orion and the Pleiades be admitted to, or excluded 
from it ? Visually, they belong to the * belt ; ' but they are 
associated with the Milky Way as well ; so that distinctness of 
constitution cannot well be granted to any stellar collection of 
which they form part. But distinctness of constitution is an 
essential feature of Dr. Gould's scheme. The adoption into 
it, besides, of a subordinate system on the scale of the Pleiades 
would imply a vaster organisation than it seems feasible to 
admit ; while its rejection, together with that of the Orion 
group, would seriously impair the integrity of the starry 
girdle forming the sensible groundwork of the theory under 

The velocities, moreover, of the stars belonging to the sup- 
posed cluster far exceed its possible gravitative power to pro- 
duce. Estimating the total mass of the system at 2,500 times 
that of the sun, and supposing it all concentrated in one point, 
the utmost speed which it could generate at the distance of 
one ' sidereal unit ' (the interval of space corresponding to a 
parallax of one second) would be three miles a second. But 
in the hypothetical cluster, this rate could practically never 
be attained, far less surpassed ; while among the stars as we 
know them, it is commonly sextupled, decupled, nay exceeded 
scores of times. It seems to follow, then, that the movements 
of the sun and stars are not interstitial movements, and that 
Gould's cluster is not organised on an independent dynamical 

Its problematical existence was entirely disregarded in 
Mr. Maxwell Hall's elaborate effort to fix the elements of the 
sidereal system. Contemplating it as an undivided whole, he 
nevertheless restricted the scope of his immediate endeavours 
to the establishment of a bond of union between the motions 
of the sun and the nearer stars. These, by his fundamental 
hypothesis, are governed by the same attractive force, and 
pursue immense orbits round a centre excessively remote even 
on the sidereal scale. Having thus laid down the broad outlines 

his scheme, he was enabled to set up certain formulae, 

cc 2 


from which, with the aid of tolerably secure data respecting a 
few stars, he arrived, by long processes of calculation, at some 
remarkable conclusions, corrected from time to time as advan- 
cing research improved the quality of the materials available.^ 

The first step was to decide upon a centre for the sun's 
movement (taken to be known in direction), and to assign 
the law of force ruling it. Motion in a circle is of course 
always aimed along a line perpendicular to the line joining 
the moving object with the centre ; hence, if the solar orbit be 
nearly circular, its centre must be removed by ninety degrees 
from the solar apex — it must be situated, that is to say, 
somewhere on a great circle of which the apex is one of the 
poles. Two diametrically opposite points on this great circle, 
one in Andromeda, the other in Hydra, gave satisfactory 
results on preUminary trial; but the choice between them 
depended upon the form assumed for the law dominant over 
sidereal revolutions. If, like those of the planets, they are 
swayed by a preponderating central mass, then the law is 
the familiar one of inverse squares, and the linear velocity, 
or speed in miles per second, of each member of the system 
decreases with increasing distance from the attractive centre. 
If, on the other hand, the movements of the stars are governed 
solely by their mutual attractions, then the sum-total of the 
gravitative power acting upon each is greater in the direct 
ratio of its remoteness from the common centre. Swiftness in 
a system thus constructed increases outwards, because angular 
velocity remains the same. All its members, in other words, 
circulate in an identical period, and will be restored, at the end 
of their annus magfuis, to their precise original positions. 

Eightly judging the latter to be incomparably the more 
probable plan of stellar organisation, Mr. Hall chose for its 
centre a spot a little to the east of e Andromedse, from which 
the velocities of certain well-known stars seemed to increase 
outward. The supposed real interval between this point and 
the sun — that is, the radius of the sun's orbit— is of no less 
than 150 sidereal units, implying a light-journey of 490 years ; 

1 Memoira B. Astr. 8oc, vol. zliii. p. 157; MontMy Notices, vols, xzxiz. 
p. 126, xlvii. p. 691. 


and the common period ascribed to the innumerable bodies 
revolving round it somewhat exceeds thirteen million years. 
Their revolutions, it may be remarked, although conducted 
(according to Mr. Hall's hypothesis) on a fundamental plan 
of almost monotonous simplicity, would present a far from 
uniform appearance. The orbits of the various stars might 
lie in any planes, be inclined at all possible angles, and be 
traversed in either direction. The sole elements of necessary 
agreement between them would be those of being described 
about a common centre in a common period. Among the 
stars coming within our range of measurement, however, there 
could, if their paths approximated to circularity, be little 
variety in real speed, since the differences of their distances 
from the supposed enormously remote centre would be in- 
significant ; and for this reason highly eccentric ellipses had 
to be assigned to exceptionally swift stars. Thus, for 1830 
Groombridge a cometary orbit (eccentricity = 0*87) was laid 
down ; moderately oval ones being appropriated to f Toucani, 
Oeltzen 17415, Oeltzen 11677, Lacaille 9352, a Draconis, 
and o^ Eridani. The conditions of movement, however, 
assumed for these bodies are totally diverse from those under 
which comets sweep through the solar system. For the stellar 
centre of force is placed, not in one focus, but at the middle 
point of ellipses traversed with velocities accelerated not to- 
wards, but away from that centre ; and two equal maxima of 
speed are accordingly attained at opposite extremities of the 
major axis. Stars moving much more rapidly than their 
fellows were then located by Mr. Hall near one or other of 
the far ends of an elongated track ; nor could the observed 
inequalities be otherwise accounted for on the basis adopted 
for his calculations. The combined mass brought out by 
them for the stellar system about equalled that of seventy- 
eight miUion suns, a result not in itself improbable, since 
Professor Newcomb, as we have seen, concluded for a maximum 
value of a hundred miUions. 

Two circumstances, however, tend to undermine confidence 
in Mr. HalFs method. In the first place, all we seem to know 
of the sun's motion in space depends upon the preliminary 



assumption that the stellar shiftings from which it has been 
elicited, include no systematic element apart from the effects 
of perspective — that deviations from the parallactic line occur 
indifferently in all directions, and are in a sense casual. But 
acquaintance with the sun's motion was an indispensable 
preliminary to the construction of Mr. Hall's scheme. The 
insecurity then of the data with which he \vorked appears to 
follow necessarily from the admission of the genuine character 
of an edifice reared upon self-destructive principles. If it 
exists, we are in ignorance as to the solar movement ; and if 
we are ignorant of the solar movement, we cannot tell whether 
it exists or not. A second objection is that the ascription of 
approximately circular or only moderately elliptic paths to 
the majority of the stars, is scarcely warranted by the analogy 
of binary stars. But if highly elliptical orbits preponderate, 
the problem of sidereal construction, as attacked by Mr. Hall, 
is practically indeterminate. 

The proffered solution of it, however, can be tried by a 
direct appeal to facts. The equations established by Mr. Hall 
gave him the means of calculating, from the known proper 
motions, the parallaxes and velocities in the line of sight of 
the stars considered. Subsequent determinations, then, in 
many cases supply a test as to the correspondence of the con- 
ditions embodied in the formulae with those existing in nature: 
The following parallaxes have been selected as among the best 
and most recently observed : — 









Gapella .... 




Procyon .... 







PoUux . 




Begulas .... 




oCygni .... 




Polaris .... 



-0-044 1 

7 Cassiopei© . 



+ 0-016 1 

^ CassiopeieB . 



-0102 1 

TOOphiuohi . 



-0046 , 




+ 0-185 

IJL Cassiopeise . 



+ 0-163 



Large discrepancies are, it will be seen, redeemed by only 
one telling coincidence ; and what traces of general agreement 
are discernible may be attributed to the effects of the inevitable 
correlation of distance to proper motion. Eadial velocities 
similarly computed show an equally distant relation to facts. 
But since one tolerably exact comparison outweighs a dozen 
loose ones, we have included in the following little table only 
the stars photographically measured at Potsdam : — 


Computal ' 



CapeUa .... 

+ 9 

+ 17 

- 8 


+ 14 

+ 30 

Polaris .... 




Procyon .... 

+ 9 

- 7 


aPersei .... 

No result 

- 7 

Rigel .... 

No result 

+ 39 

Here the sum of the errors exceeds half the sum of the 
observed quantities, to say nothing of the failure of the theory 
in two cases out of six. 

Mr. Hall cannot then be said to have overcome the diflS- 
culties encountered in his arduous enterprise ; but it must not 
be concluded thence that his labours were thrown away. The 
subject is one which, for the present, can only be approached 
tentatively ; it would be highly undesirable that investigators 
should, for that reason, be discouraged from approaching it 
at all. The life of a science is in the thought that binds to- 
gether its facts ; decadence has already set in when they come 
to be regarded as an end in themselves. ' Man is the inter- 
preter of nature ; ' to draw up an inventory, however, is not 
to interpret. It is true that speculation is prone to wander 
into devious ways ; but then ' truth emerges more easily from 
error than from confusion.' And in sidereal science especially, 
there is danger lest investigators, seduced by the wonderful 
facilities of novel methods, should exhaust their energies upon 
the accumulation of data, and leave none for the higher work 

* The sign + signifies that the star in question is receding from the earth, 
the sign — that it is approaching it. The figures given above indicate English 
iniles per second. 


of marshalling them along the expanding lines of adequate 

It is scarcely probable, indeed, that indications as to the 
general plan of the sidereal world suflSciently definite for 
purposes of numerical calculation can be gathered during the 
present era of human knowledge. A limitless field of fruitful 
research, however, lies open even now in the systems of various 
degrees of subordination, the federated combination of which 
we may reasonably suppose to constitute the supreme unity 
of the cosmos. From double, triple, multiple groupings to 
knots, drifts, clusters, clouds of stars, an ascending scale of 
complex arrangement leads upwards to the unknown — perhaps 
beyond it to the unknowable. 

It may be that no star in the entire stellar scheme escapes 
this law of particular aggregation. Even our own sun, soli- 
tary in appearance though it be, has, we are led to suppose, 
closer ties with some stars than with others, though the signi- 
ficance of the resulting movement at present escapes us. Nor 
can we be sure that a dynamical equilibrium, such as prevails 
in the planetary system, has been established in the solar or 
any other such star-group. Association of a transitory nature 
— on the cosmical time-scale — is even conceivable. One set 
of combinations may be dissolved to give place to others ; a 
single star may pass from one vast confederacy to the next, 
seeking its fortune, as it were, through space ; or, breaking 
away from the entire congeries of systems, rush out into the 
ethereal desert, to find itself, after milliards of ages, within 
the precincts of a strange galaxy beyond terrestrial ken with 
telescope or camera. 

The more attentively the configuration of the stars is 
studied, the more clearly do special phenomena of grouping 
come into view. Among the minute stars of the Milky Way, 
above all, the tendency towards arrangement in typicsl patterns 
is, in certain parts of the sky, almost as immistakable as it 
would be in a ball-room crowded with dancers suddenly arrested 
in threading their way through the figures of a quadrille 
or a minuet. Yet in the heavens, methodical distribu- 
tion must always be to some extent masked by the projection 


upon the same surface, of objects at totally different distances. 
That, under these circumstances, it should often be effaced is 
less remarkable than that it should occasionally become ap- 

One of the * typical forms ' in which stars seem to collect 
is that of an ellipse, or circle seen in perspective.* Radiated 
structures also occur; and Father Secchi, who early drew 
attention to this curious subject, regarded the presence of a 
large red star in a commanding situation among minor objects, 
as a common trait of physical groupings.* As specimens of 
a class of objects to which the ' persevering student ' could 
make large additions * with an increasing conviction as to ihe 
mutual interdependence * of their constituents, Mr. Webb singled 
out in the constellation Hercules, a * wreath,' a * double chain,' 
and a ' recurved line of small stars proceeding from one of 
7-5 magnitude.*^ The predominance of the 'spray' form 
of arrangement was dwelt upon by Mr. Proctor ; while some 
regions. Professor Holden tells us, are characterised by 
' streams ' of stars, defiling in rows of a dozen or two, across 
the telescopic field. * Others,' he continues, ' are rich in small 
definite ellipses of stars, often all of the same size. In some 
cases, stars are surrounded by circles of other fainter stars. 
In other instances, the eUipses become tolerably regular ovals, 
often of large size. These interlace in the most intricate 
manner. . . . We can frequently trace new and highly in- 
teresting features of the kind by paying attention to stars of 
one magnitude only. If we regard the eleventh-magnitude 
stars alone, for example, we may find rings and ovals of these 
stars forming a regular pattern in the sky. Interlacing with 
these may be found another pattern of similar ovals (usually 
smaller) of stars of the twelfth magnitude, and so on.' ^ 

It must, indeed, be admitted that the search for 'star- 
patterns * is over somewhat treacherous ground. Imagination 
may here easily play the part of will-o'-the-wisp. Among a 

» Century Magazine, Sept. 1889, p. 787. 

« Atti delV Accad. Pont t. vii. p. 67. ' Ceh Objects^ p. 323. 

* Monthly Notices, vol. 1. p. 62. Of. Barnard, Ibid. p. 314, Backhouse, 
Jbid, p. 374. 


multitude of scattered objects, counterfeit groupings of any 
given design can be put together by a very slight stretch of 
fancy. It has been shown experimentally by Mrs. Huggins, that 
the random distribution of dots of Indian ink (shaken from a 
brush over the paper) does not exclude the possibility, to a pre- 
disposed eye, of forming them into almost any desired figures. 
But the illustration should not, nor was it designed to lead us 
to set down as purely imaginary the visible peculiarities of 
stellar arrangement. This would be an extreme quite as mis- 
chievous as that of unqualified creduUty, since there is cumu- 
lative evidence that the peculiarities in question are, in many 
cases, real and significant. Only their recognition should be 
pursued with caution. 

The late Mr. Proctor expressed his conviction that ^ star 
streams' will eventually prove themselves genuine by the 
unanimity of their proper motions.* And it is obvious that 
they can only subsist upon this condition. Their order would 
otherwise be only a passing coincidence. Stars thus marching 
' in Indian file' are presumably swayed by an identical force 
acting upon them from a very great distance. The group 
that they form is not self-centred, but makes only a part of 
a larger organisation. Segregation, on the other hand, is the 
distinguishing * note ' of true clusters. They might be called 
autonomous democracies, each of their members obeying the 
united commands of all, while outside influences, although 
exerted upon them collectively, are without effect upon their 
internal regimen. 

A 'streaming' can then be discriminated (at least ideally) 
from a * clustering' collection of stars, by the circumstance that 
the centre of movement of the one is external, of the other 
internal. It may possibly be found that the two plans of organi- 
sation prevail respectively in diflferent sections of the Milky Way ; 
there is some appearance that they not unfrequently compete 
or combine within the same cluster, the streaming tendency 
working towards the dissolution, the clustering tendency for 
the preservation of the system. This is not the only feature 
of siderea] construction which conveys a hint to us that the 
' Universe of 8tar8t p. 84, 


world of stars and nebulae is in a state of transition. We see 
it in only one phase of its long development. To regard its 
condition as settled upon an unalterable basis, would be to 
misconstrue signs everjrwhere legible to attentive scrutiny. 

Since stars and nebulse are undeniably united into a single 
scheme, our view of the universe must embrace both classes. 
The distribution of nebulae is in fact complementary to the 
distribution of stars. Groupings of the one kind fill in the 
outlines left blank by the groupings of the other. The Milky 
Way, so far as can be discerned, is a rifted and irregular ring^ 
furnished with innumerable tentacular appendages, and com- 
posed of stars in every stage of aggregation. This ring, how- 
ever, has obvious geometrical relations with the rest of the 
sidereal structure. It marks the equator of a vast globe, of 
which the poles are canopied by the nebulae. Necessarily, too, 
of a rotating globe, since axial movement alone gives rise to 
the distinction between poles and equator. 

The opinion that the shape of the visible universe is 
spherical, or spheroidal, rather than lenticular, has been lately 
expressed by Radau,* Klein, and Falb.^ The polar relations 
of the nebulae to the plane of the Milky Way admit indeed of 
no other interpretation. And these relations can scarcely 
have been determined otherwise than by the rotation on an 
axis of the enormous, undivided volume. The condition thus 
indicated as primitively existing may have become modified 
with time ; we could, at any rate, know nothing of its preva- 
lence, since the relative situations of the heavenly bodies would, 
as M. Falb has remarked, be absolutely unaffected by it. 

All that we can see clearly is that an universal movement 
of rotation had much to do with the present distribution of 
matter in sidereal space. Whether the forces which have 
brought it about are still active must remain an open question. 
The opposite tendencies of stars to gather in the equatorial 
plane, and of nebulae to stream towards the poles of the 
system, may not even yet be exhausted ; but the point can 

Bull. Astr. t. ii. p. 88. 
' Handbuch der Himmelser$cheinungen, Th. ii. p. 812 ; Sirius, B&nde v i. 
p. 10, viii. p. 198. 


only be investigated by long and arduous methods. Their 
activity (if ascertained) would apparently imply an inherent 
difference in the qualities of the objects respectively displaying 
them; while, on the other hand, the hypothesis of stellar 
development from nebul® is indispensable, and wears the 
aspect of truth. 

We can indeed hesitate to admit neither the fundamental 
identity of the material elements of the universe, nor the 
nebulous origin of stars. The transition from one to the other 
of the two great families of the sidereal kingdom is so gradual 
as to afford a rational conviction that what we see contem- 
poraneously in different objects has been exhibited successively 
by the same objects. Planetary nebulsB pass into gaseous 
stars on one side, into nebulous stars on the other ; the greater 
nebulae into clusters. The present state of the Pleiades refers 
us inevitably to an antecedent condition closely resembling that 
of the Orion nebula ; the Andromeda nebula may represent 
the nascent stage of a splendid collection of suns. But even 
though stars without exception have sprung from nebula, it 
does not follow that nebulae without exception grow into stars. 
The requisite conditions need not invariably have been pre- 
sent. Other ends than that of star production are perhaps 
subserved by the chief part of the present nebulous contents 
of the heavens. The contrast between stellar and nebular 
distribution is intelligible only as expressing a definitive 
separation of the Ufe-histories of the two classes — a divergence 
destined to be perpetual between their lines of growth. 

The view that stars arise from a condensation of nebulous 
matter, encounters one singular difficulty. Nebulae, as we 
have seen, are devoid of sensible motion ; stars usually possess 
high, sometimes enormous velocities; and measures in the 
line of sight give it to be understood that the distinction is 
real and well-marked. It is one, however, which no theory 
of evolution hitherto broached is competent to explain ; and 
only one of two alternatives accordingly seems open to us 
— either to reject all such theories, or to assent to the in- 
ference, anomalous though it appear, that stellar movements 
are not innate, but are gradually acquired. 


Progress, then, is the law of the uniyerse. From its 
present state we can obscurely argue a * has been ' and a 
' shall be.' The face of the skies is not cast in stereotype. 
'As a vesture Thou shalt change them, and they shall be 
changed.' They shall change, by no caprice of hazard, but 
in subjection to laws unalterable in their essence although 
infinitely various in their applications, divinely directed 
towards the continually more perfect embodiment of the 
unfolding Eternal Thought. 

But the glory of the heavenly bodies, it is asserted, must 
come to an end. It results from a merely transitory state of 
things. The radiations, by virtue of which they shine, are 
the outcome of what may be figuratively termed the effort of 
nature to establish a universal thermal equilibrium. This 
condition will be attained when the frigid 'temperature of 
space ' reigns in all the millions of bodies which once were 
suns, and will thenceforward revolve, amid 'darkness that 
may be felt,' the mechanism of their movements unimpaired, 
but inert, lifeless, and invisible. Is this then the pre- 
destined end ? Science replies in the affirmative. That is to - 
say, it knows no better. Yet there is much as to which it is 
ignorant. Matter rests upon a subsensible basis, into the 
arcana of which no inquiry has penetrated. The observation 
of phenomena leads, it may be said, to the shore of an all- 
diffusive ocean of force, the existence of which is implicated 
in their occurrence. That is all we know ; at the brink of 
the ocean we pause, helpless to sound its depths, or number 
the modes of its manifestations, or predict the tasks of 
renovation or preservation committed to it. We can only 
recognise with supreme conviction that He who made the 
heavens can restore them, and that when the former things 
have passed away, and the scroll of the skies is taken out of 
sight, ' like a book folded up,' a ' new heaven and a new earth ' 
shall meet the purified gaze of recreated man. 





Name of Star 

Approxiinate Place (1890) 
« 1 B 

B Andromeds . 

7 CassiopeuB 
S GaBsiopeisB 

^ Penei . . . , 

o Ceti (Mira) . 

Pleione . . . , 

Anon. Coeli 

9' Ononis . . . , 

6^ Ononis . . . , 

U Ononis . . . . 

Lalande 13412 . 

R Geminorum . 

y ArgAs . . . , 

Anon. krgCa 

R Carinie 

R Leonis . . . . 

Arg. Gen. Gat. 14626 

Anon. Argils 

Arg. Gen. Cat. 14684 

1} Argfls . . . , 

Arg, Gen. Cat. 15177 

Arg. Gen. Cat. 15220 

Arg. Gen. Cat 15305 

8 Gentauri 

9 Mascffl . . . . 
Anon. Centanri . 

R HydriB . . . . 

fi Centanri 

Arg. Gen. Cat. 20052 

Arg. Gen. Cat. 22748 

Arg. Gen. Cat. 22763 

Stone 9168 

Arg. Gen. Cat. 23072 

Arg. Gen. Cat. 23073 

Arg. Gen. Cat. 23416 

Argelander-Oeltzen 17681 

fi LyrsB . . . . 

D M + 30°3639 

R Cygni . . . , 

DM + 48°2940 . 

X Cygni . . . . 

DM + 35°3952 . 

DM + 36°4001 . 

DM + 86°4013. 

DM + 87*>8821. 

DM + 86°8956. 

DM + 36°3987 . 

P Cygni . . . . 

DM + 88«4010. 

DM + 43«>3671 . 

V Aqnarii . . . , 

h. m. 8L 



18 13 

+ 37 


50 4 

+ 60 


1 11 34 

+ 72 


1 36 46 

+ 60 


2 13 47 

- 3 


3 42 38 

+ 23 


4 36 44 



5 29 52 

- 5 


5 29 59 

- 6 


6 49 17 

+ 20 

9-3 j 

6 49 33 



7 44 

+ 22 

62-4 ' 

8 6 19 



8 51 19 

+ 47 

100 , 

9 29 29 



9 41 39 

+ 11 


10 37 6 



10 37 15 



10 39 39 



10 40 47 



11 25 



11 1 56 



11 6 25 



12 2 40 



13 1 2 



13 11 3 



13 23 43 



13 43 


46*5 1 

14 41 49 



16 43 53 



16 44 36 



16 46 36 



16 56 21 



16 56 29 



17 11 26 



18 1 46 



18 46 1 

+ 33 


19 31 30 

+ 30 


19 33 52 

+ 49 


19 39 50 

+ 48 


19 46 20 

+ 32 


20 1 43 

+ 35 


20 6 4 

+ 85 


20 7 42 

+ 85 


20 8 6 

+ 38 


20 10 25 

+ 36 


20 12 55 

+ 37 


20 13 44 

+ 37 


20 15 31 

+ 38 


20 37 12 

+ 43 


22 19 40 








5-6 to 12-8 

F, D,, &e., bright near maximum 


Bright lines variable 

6-7 to 13 

F, &c., bright near maximmn 


White star. Hydrogen lines bright 

1-7 to 9-5 

' Bright lines near maximmn 


Thin bright hydrogen lines 

70 to 10-5? 

Bright lines near maximum 


Unknown briglit lines in photographed spectrum 


Details wanting 

G-1 to 12 

F, D, Ac, bright near maximum 


Wolf-Rayet type 

0-6 to 12-3 

Carbon bands bright near maximum 


Wolf-Rayet type 


Wolf-Rayet type 

4-3 to 9-3 

and h bright 

5-2 to 10 

Hy^liii^en llnpsi bright near maximum 


Wolf Ra vet type 


Wolf-Ka>'et type 


Wolf-Hayet t)!?© 

- 1 to 7-6 

Jki^bt hydrogen lines 


Woif Rav^t type 


Wolf-Rayet type 


Wolf Rayot typ^ 


1 Brif^bt hydroj^en lines 


Wolf-Rayet tvpe 


Wolf llftyet type 

4 to 10 

Hydrogen-lines bright near maximum 


Bright hydrogen lines 


Wolf-Rayet type 


Wolf-Rayet type 


Wolf-Rayet type 


Wolf-Rayet type 1 


Wolf^Pavet type 1 


WoU-KaVet type | 


, Wolf^Kayet type 


Wolf llayet type 

3-4 to 4-5 

Bright liues variable 


Unknown lines bright . 

5-9 to 13-0 

Hydrogen-lines bright near maximum 

7 to 11-5 

Hydrogen-lines bright near maximum | 

40 to 12-8 

Bright lines near maximum 


Wolf-Rayet type ? 


Wolf-Rayet. Cygnus No. 1 


Wolf-Rayet. Cygnus No. 2 


1 Wolf-Rayet type 


1 Wolf-Rayet. Cygnus No. 3 i 


Wolf-Rayet type ' 


1 Thm bright hydrogen lines | 


1 Wolf-Rayet type ? ; 


1 Wolf-Rayet type ' 


1 Bright hydrogen lines ^ 

D D 



( Pisoiam»2 100, 6, 8 mags, at 2d"'7. Fixed, with common proper motion. 

Photometric mags. (Harvard), 5-4, 6-4; but Webb thought large star rose at 

times to 4 mag. A close companion to B detected by Bnmham in 1888. 

Colours, both white. Speotnxm, type I. 
h 8036. Composed of two white stars never differing by above a mag., bat both 

probably variable. Binary ; in retrograde motion. 
y Arietis = 2 180. Combined mag. in Harvard Photometry 4*3. Slight relative 

variability recorded by Flammarion. Colours yellowish ; different in each 

component. Spectrum, type I. 
a Piscium » 2 202. Both stars vary in Ught, smaller in colour as well. Spectrum 

type I. 
Arietis « 2 833. Struve's estimates of mag. ranged from 4*5 to 6*5 for one, 

from 5 to 6*5 for the other component. Colours, both white. Spectrum ? 
U Tauri. Divided by Knott in 1867 into two 9*7 mag. stars. The combined 

object fluctuated considerably in light 1865-71, but has been omitted from 

recent Usts of variables. 
8 Orionis == 2 14>. Chief star varies 2*2 to 2*7. Companion, 6*8 mag. at 52"'7. 

Colours, greenish-white and white. Spectrum, type I. 
Lacaille 2145, 8"'2, 8"'5 mags. Probably binary, certainly variable (Tebbutt, 

Observatory, iv. 211). Of 6 mag. in Lacaille*s Catalogue. 
1} Greminorum varies to the extent of one mag. in a period of 229 d. Divided into 

a 3 and a 9 mag. star at 0''*96 by Bumham, Nov. 11, 1881. Spectrum, 

type III. 
S (15) Monocerotis = 2 950. Chief star varies from 4*9 to 5*4 mags, in 3 d. 10 h. 

38 m. Colours, pale green and bluish. Spectrum, type I. 
38 Geminorum « 2 982. Differences of brightness from 1*5 to 4 mags, observed 

by Strove. Colours, light yellow and purple to greenish. Spectrum, type I. 

Slow binary. 
2 1058. Companion invisible since 1844 (Burnham, Mejtioirs R, Astr, Soc. xliv. 

61 Geminorum. Mags. 5*7 to 7*5, and 9 to below 12. Distance, 60". Larger 

star deep yellow. 
U PuppisB Lalande 14551, varies 6 to 6*8 mags, in 14 d. Resolved by Burnham, 

Jan. 28, 1875, into a pair of 6*5 and 8*5 mags, at 0"*8. Colour, yellowish. 

Spectrum, type L 
2 1517. Slight alternate variability confirmed by 0. Strove. Binary. Yellowish 

y yirginis = 2 1670. Each star varies from about 3 to 3*5 mag. Colour, pale 

yellow. Spectram, type I. 
02 256 = Lalande 24098. Each star varies from 7 to about 8 mag. Spectrum, 

type II? 
YVirginiss Lalande 25086. Irregularly variable 5 to 8 mag. Composed of 

two nearly equal stars at 0"'48 (Burnham). Spectrum, type I. 

' See Nature, Tol. xxzlx. p. 65. 


re Bodtis = 3 1864. Each component varies irregnlarlj to the extent of at least 

one mag. Colours, white and ashen to yellow and blue. Spectrum, type I. 
CBootis=>2 1865. Belatiye variability noticed by F. and O. Strove. Colours, 

both white. Spectram, type I. 
2 1875. One oompo nt varies from 8*5 to 10 mag. Colours, both white. 
/3 ScorpiL Mags, estimated by Strove at 2 and 4 ; in Harvard Photometry, 3, 

5*2. Webb found a difference of 3-2 mags. ; J. Hersohel of only 1. 

Colours, yellowish-white and lilac. Spectrum, type I. 
p (6) Ophiuchi. Disparities recorded from to 2*5 mags. Colours, yellow and 

X Ophiuchi » 2 2055. Mags. 4 and 6 ; but companion varies both in light and 

colour. Large star, greenish-white. Spectrom, type I. 
2 2062. Companion at times invisible under the best conditions. 
aHerculi8=sS 2140. Large red star varies irregularly from 3*1 to 3'9 mag. 

Spectrum, type III. Green companion thought by Strove to vary from 5 to 

7 mag. Spectram shows extensive absorption of less refrangible rays. 
2 2344. Mags. 8*5 and 10 to below 12. Companion frequently invisible. 

9 Serpentis - 2 2417. Both stars seem variable. Differences from 0*34 to 1*69 
mags, registered at Harvard College in 1878. Colours, both'yellowish-white. 
Spectrum, type I. 

i3 Cygni. Large star changes very slowly from 3*3 to 3*9 mag. Colours, golden 
and azure. Spectrom, type II with modifications. 

h 1470 = Lalande 38428. Mags, estimated 7 and 8 by Secchi in 1856, but sus- 
pected to vary. Continuous observations desirable. Distance apart, 23"*8. 
Colours, * superb ' red and blue. Spectrum of larger star, type HI. 

U Cygni = Schjellerop 239a. Varies from 7*7 to below 11 mag. in 466 d. Colour, 
deep ruby. Spectrom, type IV. Companion at 62" appears to fluctuate from 

8 to 8*7 mag. in light, and in colour from blue to white or reddish. 
2 2718. Each component alternately for a short time superior. 

8 Cephei. Large star varies regularly from 3*7 to 4*9 mag. in a period of 5 d. 

8 h. 48 m. Colours, orange and blue. Spectrum, type II. 
IT. Cassiopeise = 02 (App.) 254. A similar pair to U Cygni. Bed star ( = Schj. 

280) varies irregularly from 7 to 9 ; blue star, from 8 to 10 mag. Distance 

apart, 59". Spectrum, type III. 

DD 2 




1 3 Cassiopeis 

2 Groombridge 34 

3 C Tucans 

4 a Cassiopeis . 

5 ' 1} Cassiopeiffi . 

6 fi GassiopeisB . 

7 3 AndromediB . 

8 Polaris . 

9 a Arietis . 

10 I e Eridani . 

11 I a Persei . 

12 40 (o') Eridani . 

13 Aldebaran 

^ 2S 



^* Aurigffi (oomes) 

I 51 Hev. Cephei 
' Procyon . 
1 Pollux 

10 Ursffl Majoris 

Lalande 18115 . 
I Ursaa Majoris 

Lalande 19022 
' 20 Leonis Minoris 

a Leonis . 

Groombridge 1618 

Groombridge 1646 

Lalande 21185 

Lalande 21258 . 

S 1616 

Arg. Oeltzen 11677 

S 1561 seq. 

Groombridge 1830 

Lalande 22810 . 
I a Gentauri 
! 1} Herculis 
' » Herculis 
I I'' Draconis 
I 1^ Draconis 
' Arg. Oeltzen 17415 
I 70 Ophiuchi . 
I a Lyras . 
I 1 2398 . 

a Draconis 

a Aquilffi . 

€ Cygni . 

61 Cygni . 

61 Cygni . 

a Cephei . 

c Indi 

Lacaille9352 . 

Bradley 3077 . 

85 Pegasi . 

Approx. R.A. Appnix. DccL 
181MJ 189U , 



h. m. li. 
3 18 

+ 58 32*8 



12 6 

4 43 28-9 



14 20 

-65 31-3 


2-2 . 

34 16 

1 +55 660 



42 26 

' +57 13-9 

0. Strove 


; 1 57 

+ 54 22-8 



1 3 34 

+ 35 2-2 



• 1 18 29 

+ 88 43-3 



2 58 

, +22 56-5 



3 15 30 

' -43 29-5 



3 16 28 

' +49 28-2 



4 10 13 

- 7 49-2 



4 29 36 

+ 16 17-3 



5 8 34 

+ 45 631 



6 38 48 

+ 43 41 



6 40 18 

-16 340 

Gill & Elkin 


6 48 48 

+ 87 13 

Wagner A De Ball 


7 33 33 

-H 5 30-4 



7 38 36 

1 +28 17-5 



8 53 30 

+ 42 131 

Wagner & Belopolsky 


9 7 12 

+ 53 10 



9 25 30 

+ 52 10-7 



9 36 30 

+ 43 14 



9 54 42 

+ 32 29 



10 2 31 

+ 12 30-3 

' Elkin 


10 4 46 

+ 50 1-0 



10 21 18 

+ 49 24 



10 57 20 

, + 36 56-4 



11 1 

+ 44 5-5 



11 8 3 

+ 74 4-25 

De Ball 


11 14 25 

+ 66 26-5 



11 33 

+ 45 43 



11 46 38 

+ 38 30-5 



12 4 6 

+ 41 51 



14 32 8 

-60 22-7 

Gill & Elkin 


16 39 6 

+ 39 8 

Wagner A Belopolsky 


17 11 12 

+ 36 56 

Wagner A Belopolsky 


17 30 1 

+ 55 15-6 

Wagner <^ Belopolsky 


17 30 6 

+ 55 14 9 

Wagner & Belopolsky 


17 34 4 

+ 68 27-4 



17 59 54 

+ 2 31-5 



18 33 13 

+ 38 40-9 



18 41 33 

+ 69 27-7 



19 32 34 

+ 69 28-4 



19 45 25 

+ 8 34-7 



20 41 45 

+ 88 33-5 



21 1 58 

+ 38 12-5 



21 1 58 

+ 38 12-5 

Pritchard - —-'— 


21 16 

+ 62 7 



21 54 56 

-57 14-3 

Gill (& Elkin 


22 58 46 

-36 29-2 



23 7 56 

+ 56 33-6 

Gyld^n - 

5-8 1 

23 56 26 

+ 26 30-1 

Briinnow - . 








Dist«nce iu 


In Miles per 
























2 1 

























; Photography 

















1-4 1 






71-5 ' 






4-8 , 












? i 







Meridian circle 











13-9 1 







Meridian circle 

. 1889 





Meridian circle 






Meridian circle 






Meridian circle 






Meridian circle 


















Meridian circle 






























Meridian circle 













— ^Meridian circle 







— JHeliometer 






.3feridian circle 






m^eridian circle 






Meridian circle 






^ Meridian circle 







_- Heliometer 






_^ Heliometer 






. _ Heliometer 












— Micrometer 






__ Heliometer 






- Photography 













— Photography 












- Heliometer 






— Heliometer 






^ Micrometer 


0-283 1 







0-054 1 






(Kindly cammimieaUd by Profesior E. SchOnfdd of Bonn,) 

Name of star 
Groombridge 1880 



DeoL (185S) 




h. m. 8. 
11 44 86 

+ 88 45-5 


B3. vii.. 112= LL 

Laoaille 9852 


22 56 29 

-86 41-2 


Cordoba 82,416 . 


28 56 51 

-88 4-2 


61 Cygni . 


21 24 

+ 88 2-3 


dnpl. 17' 1 

Lalande 21185 . 


10 55 24 

+ 36 66-4 


B.B. vii., 104 

c Indi . 


21 52 14 

-57 22-5 


Lalande 21258 . 


10 58 15 

+ 44 16-2 


B3. vii., 105 

40Eridani . 


4 8 86 

- 7 62-9 


tripl. (comes dapL i 
4", dist. 82") 

fi Cassiopeias 


58 89 

+ 54 12-7 


a Centauii . 


14 29 48 

-60 18-9 


dupl. 15" 

Aig.Oeltz. 14818. 
Arg. Oeltz. 14320 . 


15 2 17 
15 2 17 

-15 45-9 
-16 40-9 


di8t.= 5'-0 

LaoaiUe 8760 


21 8 44 

-39 26-5 


e Eridani 


3 14 7 

-43 37-6 


Arg. Oeltz. 11677 . 


11 12 27 

+ 66 37-9 


Oroombridge 84 . 


10 7 

+ 43 120 


dupl. 40" 

Piazzi II., 123 


2 28 8 

+ 6 11-5 


Tialande 25372 . 


13 38 23 

+ 15 40-7 


a Bootis 


14 9 8 

+ 19 56-8 


J 2398 . 


18 41 10 

+ 59 24-5 


dnpl. 16''. Strove 
p.m. 2164 

3 Hydri 


18 4 

-78 4-3 


Lalande 7448 


8 53 31 

+ 34 551 


WeiBse V. 592 


5 24 7 

- 3 42-3 


1 Bradley 8077 


23 6 19 

+ 66 221 


C ToucansB . 


12 28 

-65 43-7 


Lalande 15290 . 


7 44 17 

+ 31 2-9 


PiaKzi XIV., 212 . 


14 49 

-20 45-5 


dnpl. 13". B. A. C. 

rCeti . 


1 87 20 

-16 42-6 


Laoaille 661. 


2 4 85 

-51 32-2 


<r Draconis . 


19 82 38 

+ 69 24*8 


Weisse IX. 954 . 


9 43 56 

-11 85-7 


Fedcwenko 1457. 


7-7. 7-9 

9 4 29 

+ 68 18-3 


dupl. 20". LI. 18116 

Lalande 80694 . 


16 45 41 

+ 16-7 



8 Pavonis . 


19 54 26 

-66 82-6 


Lacaille 8362 


20 1 40 

-86 27-8 


LacaiUe 2957 


7 40 11 

-83 53-3 


61 Virginia . 


13 10 49 

-17 80'2 


Lalande 31056 . 


16 57 81 

- 4 490 


, Piazzi h. 180 . 


29 54 

-26 33-9 


' 20 Mayer . 
Lalande 6888 


40 47 

+ 4 82-0 


B. A. G. 221 


3 87 9 

+ 41 1-2 


Fedorenko 1884 . 


8 41 40 

+ 71 21-0 


Groombridge 1618 


10 2 28 

+ 60 111 


Lalande 30044 . 


16 23 21 

+ 4 33-2 


Lalande 38383 . 


19 57 49 

+ 22 58-3 






Name of Star 


B. A. (1866) 




h. m. s. 

O / 


V Indi . 


22 12 5 

-72 57-8 


Lalande 46650 . 


28 41 38 

+ 1 380 


c Persei 


2 58 88 

+ 49 8-7 


WeisselV. 1189 . 


4 53 37 

- 5 55-6 


Sirius . 


6 38 46 

-16 81-8 


dupl. 10" 



7 31 43 

+ 5 85-6 


TiACAille 3386 . 


8 27 12 

-31 2-4 


LacaillA 4887 


11 39 36 

-89 42-6 


Lalande 27744 . 


15 6 85 

- 47-2 


y Serpentis . 


15 49 46 

+ 16 8-3 


Arg.Oeltzen 17416 


17 87 17 

+ 68 28-3 


Pia7.7.i XX. 29 


20 6 14 

-27 27-7 




23 9 34 

-14 86-6 


dupl. 1" (Bun 

dupl. 0-"5 (Bun 

dupl. 7" 


85 Pegasi . 


23 54 43 

+ 26 18-9 



11 GassiopeiaB 


40 21 

+ 67 2-8 


5 Trianguli . 


2 8 14 

+ 33 33-5 


Lalande 16565 . 


7 61 33 

+ 29 39-0 


Lacaille 4966 


11 60 45 

-26 52-5 


43 Coma 


13 5 7 

+ 28 36-9 


Lalande 28607 . 


15 36 18 

-10 27-3 


Weisse XVI. 906 . 


16 47 45 

- 8 3-8 


36 Ophiuohi . 


17 6 27 

-26 22-1 1 



= 12'-1 

30 Scorpii . 


17 7 18 

-26 19-0/ 

Weisse XVn. 322 


17 18 33 

+ 2 17-5 


Lamontj 3744 


18 50 55 

+ 6 46-0 


Arg.Oeltzen 20452 


20 15 2 

-21 47-4 


f»Reticuli . 


3 14 36 

-63 7-9 \ 
-63 3-7/ 



C Retiouli . 


3 15 1 

LacaiUe 2138 


5 48 48 

-80 34-8 


e Ursae Majoris . 


9 23 8 

+ 52 20-1 


20Grateri8 . 


11 27 27 

-82 3-8 


Lalande 27298 . 


14 51 4 

+ 64 14-9 


70 Ophiuchi 


17 58 8 

+ 2 32-3 


dupl. 4" 

Lalande 16804 . 


8 11 81 

-12 8-6 


Tialande 27026 . 


14 43 27 

-23 41-1 


w Herculis . 


17 15 8 

+ 32 39-4 


b AquilsB . 


19 18 14 

+ 11 38-3 


Lacaille 8620 


20 48 2 

-44 38-6 





I ^ 

■' 2 1 
2 s I 
'I 1 s- 

aa «« g 


9 S 
I I 




1 .2| .3 J 

o g> g s<.r 










X = mnii«{os 

Ill999i8 JO tBVK 

ani JO spaooM 


04 fH kb r- 

C4 iH «0 C9 

00 O 

O lO o 

6 A. 6 


OS OS e- 

to o 


mil i 



Tabj^ VI.— movements IN LINE OF SIGHT 

(The signs + and — signify respectively recession from, and approach 
to, the earth). 


Name of Star 

Photographically measured at Potsdam 

measured at 


+ 30 English miles a second 

+ 31 

Capella . 

+ 17 „ 

+ 23 


+ 39 1, „ „ 

+ 18 

€ Ononis . 

+ 34*6 „ „ „ 

+ 15 

Polaris . 

- 16-3 

o Virginis 

~" 14 11 »» f» 

- 17 

o Persei . 

- 7 „ 

- 22 

Procyon . 

~ * »» »» »i 

+ 3 

Algol . . 

— 2*3 „ „ „ 

Arcturus . 

- 45 

Vega . . 

- 34 

a Cygni . 

- 36 

Pollux . . 

- 33 1 

a Aqnilae . 

-27 , 

a Ononis . 

+ 28 1 



Abbe, distribution of nebulsB, 374 

Aberration of light, annual, 299, 330 
secular, 329-30 

Abipones, their Pleiad-ancestry, 221 

Abney, amassing power of sensitive 
plates, 23; scale of photographic 
density, 29 ; analogy between 
spectra of Sohag prominence and of 
a AqoilsB, 47 

Airy, Sir Gteorge, stellar parallax, 300 ; 
solar translation, 324 

— Miss, twelve Pleiades seen by, 223 

Albategnius, red stars, 146 

Albireo. See Cygni, jS 

Alcyone, unnoticed by Ptolemy, 222 ; 
proper motion, 224, 225; distance 
from the earth, 226; intrinsic 
brilliancy, 226, 227; movement of 
approach, 229; connected nebuls, 
230, 231 ; primary in group, 234 

Aldebaran, magnitude, 3 notet 21, 22 ; 
chemical composition, 46 ; colour, 
46, 146, 151 ; attendant star, 175 ; 
parallax, 314 ; recession from the 
earth, 341, 391 

Alexander, Owl nebula, 256; spiral 
theory of Milky Way, 862 

Algol, spectrum, 87, 137 ; light- 
changes, 130, 185-6; spectroscopic 
measurement of motion, 137; 
system formed by, 138, 148 note, 
174 ; colour, 147 

Algol-variables, 130, 135-144, 174, 

Al-SM, colour of Algol, 147; of 
a HydrsB, 147, 151 ; observations of 
Y Virginis, 182; of the Pleiades, 
222 ; of Andromeda nebula, 268 

Altair, appellation, 2 ; magnitude, 8 
notCy 21, 22 ; spectrum, 42, 47, 80 ; 
parallax, 814 

Andromedffi, 7, colours, 157, 206 ; a 
triple star, 206 

Angstrdm, relative age of the stars, 85 


Antares, magnitude, 8 note; banded 
spectrum, 53 ; efficiency as a 
radiator, 61 ; ruddy tint, 146, 151 ; 
green satellite, 156, 176 

Anthelmus, discovery of new star, 98 

Apex, of the sun's way, 321-2, 323, 
324, 326, 328, 386, 388 ; must shift 
with time, 329 

Aquarii, C> period of revolution, 195 

AquilfB, a. See Altair 

— 17, light-changes, 138, 184 

Arcturus, identified by Homer, 2 ; 
magnitude, 3 note, 22 ; spectrum, 
47, 80, 346 ; stage of development, 
49, 91; colour, 146, 147, 151; 
parallax, 805, 314, 390; velocity, 
315, 845; spectroscopicaJly deter- 
mined, 341, 342 

Argelander, stellar enumerations, 1, 
6, 11, 837, 364, 365 ; place of 
Tycho*s new star, 98 ; nomenclature 
of variables, 100; periodicity of 
Mira, 112; of jS Lyraa, 130; of 
S Gancri, 139 ; colour of Arcturus, 
147 ; scale of magnitude, 337, 865 ; 
motion of 1830 Groombridge, 840 ; 
distribution of stars in Milky Way; 

Argills, 7, spectrum, 70-1, 72 ; situation 
in Milky Way, 354 

— 17, spectrum, 70, 120 ; colour, 
116, 117; situation in a nebula, 
116, 282 ; light-changes, 116-119 ; 
relation to Milky Way, 854 

Arietis, 7, discovery as a double star, 
168 ; slowness of revolutions, 176, 
195 ; variability, 182 

Atlas, one of the Pleiades, 228 ; light- 
power, 226; supposed companion, 

Atmospheres, stellar, absorption by, 
35, 60, 58. 85. 148, 161, 152, 161 ; 
extent, 61, 68, 90, 127, 150 ; relation 
to Ught-change, 109, 122, 127^8 




Aarigffi, /9, speotrographic discovery of 

binary character, 213 
Aurorte, scintillation of stars daring, 

Australian tribes dance in honour of 

the Pleiades, 221 ; specimen of 

their stellar folk-lore, 8.'i5 
Auwers, new star in Scorpio, 103; 

orbit of SiriuR, 172 ; nebulous triple 

Star, 216 ; stellar parallax, 303, 300 ; 

reduction of Bradley's observations, 

324; proper motions of stars in 

Ursa Major, 347 

Bailet, photographic spectrum of 
il Argi\s, 120 

Bail, L. de, solar translation, 32G 

— Sir Bobert, parallax of Nova Cygni, 
105 ; of 61 Cygni, 303 ; search for 
large parallaxes, 317 

Banded stellar spectra, 52-65 ; third 
type, 53 ; fourth type, 62 

Barlow, <r Ononis, 215 

Barnard, detection of a star in the 
trapezium, 218 ; photographs of 
Btar-groups, 237, 273, 356; dis- 
covery of nebulous ring, 249, 262 

BaxendcU, variations of U Tauri, 183 

Bayer, designation of stars, 3, 110, 113, 
207, 244 ; Andromeda nebula un- 
noticed by, 268 

Becker, bright-line spectra, 68, 70, 

Bessel, Bradley's observations, 12; 
observation of P Cygni, 70 ; dis- 
turbed motion of Sirius, 171; 
common proper motion, 174 ; mea- 
surements of Pleiades, 224, 226, 
227 ; stellar parallax, 300, 302-3, 
304, 805; partiid stellar system, 

Betelgeux, designation, 2 ; magnitude, 
8, 22; spectrum, 53, 54, 57; 
chemical composition, 54, 56 ; stage 
of development. 86, 88, 90; 
variability in light, 120 ; colour, 
146, 151 ; negative parallax, 314 

Binary stars, 166. See Stars, double 

Birmingham, new star, 99 ; red stars, 
150, 151, 162 ; distribution of, 164 

Blue stars, 160, 166, 162, 176, 186 

Boeddicker, delineation of Milky Way, 
856, 363 

Bond, G. P., double star photography, 
190 ; nebulous triple star, 216 ; 
observations of Andromeda nebula, 
269, 270, 272; structure of Orion 
nebula, 278 

Bodtis, 44 ; variable double star, ISO ; 
period of revolution, 188 

— C) variable double star, 180 

BouiUaud, phases of Mira, 110; 
Andromeda nebula, 268 

Bradley, observations of star-places. 
12, 324. 334, 337 ; observations of 
Castor, 166 ; of 61 Cygni, 168 ; of 
7 Arietis, 176 ; stellar parallax, 298 
aberration of light, 299 

Bredichin, parallax of a nebula, 257 

Brester, theory of stellar variability. 

Bruce telescope, 26, 27 

Briinnow, parallax of a nebula, 267 ; 
stellar parallax, 308 

Brunowski, star of 1604, 98 

Buffham, dark lanes in Hercules 
cluster, 246 

Burchell, brightening of v Argiis, 117 

Bumham, new double stars, 165, 205 ; 
Y Virginis, 183 ; ij Geminorum 184 ; 
fi Delphini, 194; jS Scorpii, 209; 
y Scorpii, 214; <r Ononis, 215 ; ob- 
servations of Sirius, 172 ; of 
Procyon, 173; of Rigel, 206; 
variable satellites, 182, 186 ; stars of 
trapezium, 217; double stars in 
Pleiades, 227 ; nebulous star, 253 

Burton, spectrum of a southern 
nebula. 81 

Calciuii-lznes in stellar spectra, 

Cancri, S, Algol-variable, 136 ; nature 
of light change, 139, 143; light- 
curve at minimum, 140, 141, 144 

Cancri. f, changes of colour, 159 ; pe- 
culiarities of system, 209-211 

Canes Venatici, spiral nebula in, 263, 

Canis Majoris, B. Algol variable, 136, 

Canopus, magnitude, 3 ; spectrum, 87 ; 
real brilliancy. 310 

Capella, magnitude, 3, 22; physical 
resemblance to the sun, 46, 47, 49 ; 
stage of development, 91 ; colour, 
147 ; parallax, 306, 390 ; direction 
of apparent movement. 821 ; 
speotrographic determination of 
movement in line of sight. 348. 

Cape of Good Hope, heliometer, 16, 
312; photograph of comet, 24; 
photographic star-charting, 7, 381 
noU ; stellar parallax, 30a-10, 312, 
317. 349 ; star-catalogue, 326 




Carbon, suggested presence in stars, 
56-7, 71, 105; undoubted spectro- 
scopic effects, 62-3, 74 

Garrington, fourteen Pleiades seen by, 

Cassiopeia, new star in, 97-8, 154 

CassiopeiaB, 7, peculiarities of 
spectrum, 66-7, 72 ; attempts at 
physical explanation, 68-9 

Cassiopeia), 77, colours of components, 
167, 170 ; revolutions, 171 ; masses, 
196, 199 

— /It, intrinsic light-power, 315, 346 ; 
velocity, 316, 345 

— U, redness set off by contrast with 
blue attendant, 152, 184 

Castor, spectrum, 37, 91 ; discovery 
as a double star, 163 ; revolutions, 
166, 192; remote satellite, 176, 

Catalogues, stellar, 12, 324, 326, 334, 
337 ; photometric, 21 ; photo- 
graphic, 26, 381 note; spectro- 
graphic, 30; of variables, 108; of 
red stars, 150-1 ; of double stars, 
165, 170, 348 ; of Pleiades, 223 ; of 
nebulsB, 235 7iote, 265 ; of stars in 
various clusters, 238, 240, 241 

Celoria, stellar orbits, 191, 194; star- 
soundings, 366, 383 ; compound 
Milky Way, 367 

Centauri, a, lustre, 3 ; spectrum, 44 ; 
noted as double, 163 ; revolutions, 
167-8, 197 ; mass, 168, 195 ; dis- 
parity between mass and light of 
components, 199, 211 ; their rela- 
tive radial velocities, 201 ; parallax, 
302, 308 ; attractive power over the 
sun, 320 ; proper motion, 340 

Centauri, a>, globular cluster, 244 

Cephei, 5, light changes, 132, 143, 183 ; 
contrasted colours of componentn, 
157, 176 ; mutually fixed, 195 

Cephei, B, Ught-changes, 119, 121 

Cephei, U, Algol-variable, 136 ; cha- 
racter of variations, 141 

Ceraski, variability of U Cephei, 141 

Chacornac, variable nebula, 293 

Chambers, red stars, 150, 151 

Chandler, catalogue of variables, 108 ; 
colours and periods, 127 ; perio- 
dicity of Algol-stars, 141, 142 

Chemistry, of stars, 36, 42, 46, 54, 
56-7, 62-3, 64, 67-8, 71 ; of nebulae, 
75-6,77, 81 

Chromosphere, solar, 44, 49; in 
Sirian stars, 51 

Cicero, colour of Sirius, 146 

Clark, Alvan G., satellite of Sirius, 

172; detection of a star in trape- 
zium, 218 

Cluster, M 80, new star in, 103-4 ; 
about fc Cruois, 165, 240; in 
Gemini, 236; bifid, 236; Mil, 
236-8 ; double, in Perseus, 239-40, 
331, 376; globular, in Hercules, 
242, 246, 247-8, 249, 250, 282 

Clusters, red stars in, 154, 239, 240 ; 
nebulous affinities, 219, 232, 250, 
281, 296; defined, 220; irregular, 
234-41; globular, 241-8; spectra, 
250; galactic relations, 373, 377; 
distribution, 374, 376, 377 

Coal-sack, in Milky Way, 355, 356, 
301, 362 

Coit, elements of tj Cassiopeia, 171 

Colours, of double stars, 29, 155-162, 
167, 168, 170, 172, 176, 180, 182, 
183, 184-5 ; dependent upon atmo- 
spheric absorption, 37, 53, 63, 85, 
148, 161 ; of Sirian and solar stars, 
37, 44 ; of stars with banded 
spectra, 53, 63; of red stars, 85, 
146-7, 150-4 ; of temporary stars, 
97, 99, 100, 105, 154 ; of variables, 
108, 116, 117, 121 ; connection with 
period of light-change, 127 ; varia- 
tions, 152^, 158-60, 176, 180, 182, 
185; of clustered stars, 239, 240, 
245 ; of nebulffi, 254, 256, 257, 258, 
259, 278 

ComaB Berenices, 42, apparent orbit, 
188, 194 

Comet, of 1882, photographed, 24; 
envelopes, 287; of 1861, 287; 
Halley's, 262, 289; Sawerthal's, 
287 ; Pons's, 289 

Comets, chemical analogy with 
nebulflD, 76, 77, 285 ; structural, 262, 
285-9; collisions with stars, 107; 
in stellar systems, 168 

Common, nebular photography, 32 ; 
Nova Andromedae, 104 ; nebula? in 
Pleiades, 230, 231 ; cometary nebula, 

Constellations, origin, 2 

Copeland, spectrum of y Aigds, 70 ; 
discoveries of gaseous stars, 71-2 ; 
spectra of Orion nebula, 78, 81, 101 ; 
of Nova Cygni, 102 ; of Nova 
Andromedae, 105 ; U Geminorum, 
116 ; red star, 152 

Copernicus, distances of the stars, 297 

Cor Hydrae, red star, 147, 151 

Cornu, ultra-violet spectra of 
hydrogen, 38; of thallium and 
aluminium, 39 ; spectrum of Nova 
Cygni, 100 




Corona, solar, calcium-rays in spec- 
trum, 45; nature, 50; analogous 
appendages to stars, 51, 61, 62; 
changes of emissive intensity, 126 

Coronie, B, irregular extinctions, 120 

— T, new star, 100, 103, 106. 154 

— U, Algol variable, 136, 141 
Goronium, 60 

Croll, collisions of stars, 94 

Cross, Southern, situation in Milky 

Way, 365, 357, 366 
Cmcis, a, magnitude, 3 ; spectrum, 37 ; 

duplicity. 163, 175 

— 7, spectrum, 53 ; proper motion, 

— «, cluster about, 155, 240 
Cruls, cluster about k Crucis, 240 
Cyclones, announced by the twinkling 

of stars, 6 
Cygni, a, magnitude, 3 ; spectrum, 42 ; 
negative parallax, 305, 314, 390 ; ap- 
proach to the eartli, 341 

— fij contrasted colours, 157, 175; 
spectra of components, 161 ; noticed 
as double, 163 ; a fixed pair, 176, 
195; light-change, 183 

— 5, colour-changes of satellite, 160, 
182 ; discovery, 181 ; photographic 
method inapplicable, 191 ; mass- 
briUancy, 203 

— Xi spectrum, 70, 114 ; light- 
changes, 113-4, 125 

— 61 ; duplicity detected, 163 ; 
proper motion, 168, 175, 340 ; orbit, 
169-70, 192; mass, 169, 195; in- 
trinsic lustre, 170, 203, 304 ; light in- 
equality, 202 ; parallax, 302-3, 305, 

815 ; situation in Milky Way, 328 

— P, history, 70, 99 ; spectrum, 222 

— U, contrasted colours, 152, 184 

— y, Algol- variable,136, 142 

Dabk rifts, in clusters, 236, 246, 282 ; 
in nebulsB, 282, 283 ; in Milky Way, 
363-4, 855-6, 361, 362-8 

Darquier, discovery of ring nebula, 260 

D'Arrest, Tycho's star, 98 ; red star, 
152 ; conformation of a cluster, 235 ; 
nebulous star, 258; spectrum of a 
nebula, 254 ; nebular immobility, 
267 ; absence of nebulous light in 
Orion trapezium, 277 ; satellites to 
nebulae, 291 ; variable nebuliB, 292, 
293, 294 

Dawes, colours of double stars, 156, 
159, 160 

Declination, defined, 18 

Dc la Bue, reflecting telescope, 32 


Delphini, 7, colours, 158, 160; slowness 
of revolutions, 195 

Dembowski, colours of double stars, 
159, 160; variability of a pair in 
Virgo, 181 ; duplicity of U Puppis, 

Democritus, constitution of Milky Way, 

Deneb. See Cygni, a 

Denning, number of Pleiades seen by, 

Density, low, of Algol -stars, 138, 140, 
142. 143 ; light of double stars rela- 
tive to, 193, 202-3, 208, 211 

Derham, observation of a cluster, 237 

De Thou, early star-map showing 
Andromeda nebula, 268 

Differential method of parallaxes, 801 

Discs, spurious, of stars, 2 

Disc-theory of Milky Way, 853 

Distances, of the stars, 105, 144, 297- 
315 ; of the Pleiades, 226 ; of Milky 
Way collections, 364-6 ; relative, of 
clusters, 246, 374 ; of nebulie, 257, 
369-71, 374-5, 377-8 

Distribution, stellar, 19, 27; of red 
stars, 65, 104 ; of gaseous, new and 
variable stars, 74, 106, 144-5 ; of 
nebula) and clusters, 371-8, 395 

Doberck, stellar orbits, 191, 192, 195, 
203, 208 

Dollen, stellar parallax, 308 

Dreyer, red star, 152 ; catalogue of 
nebulie, 104, 235 note; movements 
of nebulie, 290 ; variable nebula, 293 

Dun^r, banded stellar spectra, 55 ; dis- 
tribution of fourth-type stars, 65 ; 
spectra of variables, 122 ; red st^rs, 
152, 153 ; double-star observations, 
180 ; colours of 5 Cygni, 182 ; stellar 
orbits, 191 

Durchmusterung, Argelander*s, 6, 11, 
100, 837, 864, 866 ; Schonfeld's. 7 ; 
Giirs, 7, 381 noU. 

East&ian, stellar motions and magni- 
tudes, 339 

Eccentricity, of binary star orbits, 139, 
204, 390; of orbits in sidereal 
system, 389, 390 

Eclipses, of Algol -variables, 136-148 

Electra, magnitude, 223 ; light-power, 
226 ; movement, 228 ; nebulous ap- 
pendages, 230, 231 

Electrical condition of nebulfe, 88, 289, 

Electrical repulsion in sun and stars, 
60-1, 61 



Elements, ohemioal, reoognised in 
stars, 87. 42, 44, 46-7, 64, 60, 62. 
68, 70 ; in nebnlsB, 77 

Elkin, heliometer used by, 16, 225 ; 
measurements of the Pleiades, 225, 
226, 228; parallaxes of sonthem 
stars, 308-10 ; mean parallax, 813 ; 
parallaxes of northern stars, 814-5, 

Engelmann, Nova Andromeds, 104 ; 
colours of 8 Cygni, 160 

Eqnatoreal telescope, 15, 24 

Eqaulei, 5, orbit, 194 ; brilliancy, 203 ; 
optical attendant, 213 

Equulei, e, triple system, 208 

Eridani, o^, mass, 195 ; feeble lumi- 
nosity, 203, 207 ; ternary group, 207 ; 
parallax, 309 ; orbit in sidereal 
system, 389 

Eridani, e, parallax, 309, 390 ; proper 
motion, 310 

Eridani, 6, discrepancies of, observed 
with computed movements, 191 

Espin, banded spectra, 53, 63 ; bright 
lines in stellar spectra, 60, 114 ; 
spectrum of U Geminorum, 116 ; 
light change of 63 Cygni, 119 ; sub- 
class of variables, 120 ; fluctuations 
of a red star, 121 ; of U Puppis, 184 

Evolution, sidereal, 82-5, 88, 92-4 ; of 
double stars, 170, 199, 204; of 
clusters, 219, 273-4, 281, 396; of 
nebulae, 369, 396 

Exner, twinkling of stars, 6 

Extinctions, method of, 21 

Fabricics, discovery of Mira, 110 
Falb, form of the visible universe, 895 
Farley, colour of R Cephei, 121 
Feuill^e, observation of a Centauri, 167 
Finlay, magnitude of i) Argills, 117, 120 
Flammarion, colours of double stars, 

159, 176; variability of y Arietis, 

182 ; movements of i Gancri, 210 ; 

Omega nebula, 282 ; quadruple drift- 
ing group, 348 
Flamsteed, designations of stars, 207, 

213 ; nebulous star, 253 ; illusory 

parallax, 298, 299 
Fluted spectra, 52, 53, 68. 76, 99 
Fomalhaut, magnitude, 3 ; movement. 

Frankland, spectrum of hydrogen. 69 
Franks, star-colours, 149, 159 
Franz, parallax of Nova Andromedse, 

Fraunhofer, lines in solar spectrum, 

39. 44 ; improvement of telescopes, 

301 ; heUometer, 304 


Galilbo, distance of the stars. 297 ; 
constitution of the Milky Way, 362, 

Galloway, translation of solar system, 

Gambart, observation of 8 Cygni, 

Geminormn, (, periodicity, 133 

Geminorum, 17, variable double star, 

Geminorum, B, remarkable spectrum, 

Geminorum, U, light-changes. 115. 
124 ; spectrum, 116 

German Astronomical Society, deter- 
mination of star-places, 12 

Gill, photographic registration of stars, 
7, 381 note ; parallaxes of southern 
stars, 308-10 ; problem of star- 
distances, 311 ; Dunecht heliometer, 

Gilliss, southern zones, 12 ; observa- 
tion of 9} Argi!is, 117 

Glasenapp, stellar orbits computed by, 

Goldschmidt, nebula about Pleiades, 

Goodricke. variability of iS Lyrse, 130 ; 
of Z Cephei, 182 ; of Algol, 136 

Gore, catalogue of variables, 108 ; U 
Ononis, 113 ; 10 Sagittas, 133, 143 ; 
elements. of stellar orbits, 171, 173, 
182, 191, 194, 207 ; galactic streak, 
357 ; star-distribution, 382, 384 

Gothard, nebular photography, 32 ; 
bright-line spectra, 67 ; photographs 
of ring-nebula, 260, 261 ; of spirals, 
263-4; of comet-like nebulee, 28()- 

Gould, stars seen at Cordoba, 1 : zones, 
12 ; light-changes of southern stars, 
133, 134; red stars, 152, 153; 
measurement of a photograph of 
the Pleiades, 224; swift southern 
stars, 340 ; solar cluster, 384-5, 386, 

Greenwich, spectroscopic determina- 
tions of motion, 229, 341, 343 ; 
Bradley's observations at, 824 

Groombridge, ciroumpolar catalogue, 

Groombridge 1880, rapid movement. 
826. 340, 344, 345; light-power, 
346 ; eccentricity of ascribed sidereal 
orbit, 389 

Gully, observation of Nova Andromeda;, 

Gyld^n, theory of stellar variability, 





Hahn, central star in ring-nebula, 261 

Hall, A., parallax of Nova Andromedie, 
105 ; 0- Ononis, 215 ; stars in 
Prssepe, 241; stars about ring- 
nebula, 260 

Hall, Maxwell, Merope nebula, 230; 
sidereal organisation, 387-91 

Halley, observations of nebulas, 10; 
V Argds, 116. 117; clusters, 237, 
247 ; comet, 262, 289 

Harding, ring-nebula in Lyra, 260 

Hartwig, Nova Andromedaa, 104 

Harvaid College, photometry, 21 ; 
photographs of stellar spectra, 30, 
57-8, 64, 71; periodical e£Fects of 
motion, 212-13 ; new nebulie, 281 

Heis, stars visible at Monster, 1 ; 
maxima of Mira, 110 

Heliometer, 15, 304 ; Gape, 16, 312 ; 
Konigsberg, 224, 306 ; Yale College, 
225, 311 ; Bonn, 306 

Helium, a solar element, 60 ; charac- 
teristic line bright in stellar spectra, 
60, 66, 70, 101. 102, 113, 114; in 
spectrum of Orion nebula, 78 

Helmert, clustered stars in M 11, 238 

Henderson, parallax of a Centauri, 

Henry, Paul and Prosper, photographic 
telescope, 25 ; photographs of double 
stars, 190; of Pleiades, 223, 231, 
232 ; of cluster in Gemini, 235 

Hercules, cluster in, radiated structure, 
243, 247 ; photographs, 246, 249 ; 
dark lanes, 246, 282 ; mass of com- 
ponents, 248 ; spectrum, 250 

Herculis, a, spectrum, 53, 129 ; chemi- 
cal composition, 54 ; light-changes, 
120, 129, 183; colours, 157, 183; 
slowness of revolutions, 195; Bpu- 
rious parallax, 307 

Hp.rculis, C« period of revolution, 194 ; 
spectroscopic parallax, 301 

Herculis, 95, colour-changes, 159, 160, 
176 ; slowness of revolutions, 195 

Herschel, Caroline, attendant upon 
Andromeda nebula, 269 

Herschel, Sir John, constellations, 2 ; 
outburst of ri AigdSf 117; red stars, 
148, 151, 162 ; colours of y Leonis, 
158 ; of a Centauri, 167 ; of stars 
about K Crucis, 240 ; of 47 Toucani, 
245; measure of a Crucis, 175; 
observation of 8 Cygni, 181 ; system 
of C0ancri,-210; nebula; and double 
stars, 216 ; sixth star in trapezium, 
217; rifted clusters, 236; observa- 
tions of clusters, 237, 238, 244, 245, 
246, 377; dynamics of globular 

cloBters, 241-2 ; nebulous star , 
253; planetary nebuUe, 253, 255, 
258 ; satellites to nebulie, 260, 291 ; 
annular nebulie, 261, 262 ; falcated 
nebuls, 262 ; M 51, 263 ; doable nebu- 
lie, 265, 266; catalogue of nebulae, 
265 ; Dumb-bell nebula, 266 ; ellip- 
tical nebulflB, 275, 276 ; Orion nebula, 
278; Argo nebula, 282; Omega 
nebula, 283; Trifid, 283-4, 295; 
variable nebula, 293 ; parallax of 
61 Cygni, 303 ; galactic relations of 
stellar movements, 350; observa- 
tions of MUky Way, 354, 355, 357 ; 
an annular structure, 361 ; nebular 
regions, 371 ; nubeoulfls, 370, 372 ; 
number of the stars, 381 ; zone of 
large stars, 384 

Herschel, Sir William, nebular dis- 
coveries, 10; planetary nebulie, 77, 
255, 259 ; nebulous origin of stars, 
84 ; maximum of Mira, 110 ; varia- 
bility of a Herculis, 120 ; red stars, 
148 ; colours of double stars, 155, 
158, 159; double-star discoveries, 
164, 181, 192, 203, 207, 208, 209, 
213 ; 0- Ononis, 215 ; arrangement 
of stars in clusters, 235, 236, 238, 
247 ; cluster in Perseus. 239 ; 
globular clusters, 241; forms of 
nebulff, 251 ; nebulous stars, 252-3, 
254; network nebula, 277; vast 
nebulous region in Orion, 281 ; 
Omega nebula, 282 ; diffused 
nebulosities, 284 ; nebulie and 
comets, 285 ; variable nebulae, 293, 
294 ; translation of solar system, 
321, 322, 323, 386; systematic 
parallax, 332; observations of 
Milky Way, 355, 357 ; star-gauges. 
358-9, 365, 382, 383; clustering 
effects, 359-60, 864; vacancy in 
Scorpio, 361 ; nebular distribution, 
371 ; construction of the heavens, 

Hesiod, stars known to, 2; heliacal 
rising of the Pleiades, 221 

Hevelius, designation of Mira, 110; 
observation of R Cephei, 121 ; 6 
Orionis singly catalogued by, 217 

Higgens, colour-change of 95 Herculis, 

Hind, red stars, 63, 151, 153 ; Tycho's 
and Anthelm's stars, 98 ; Nova 
Ophiuchi, 99 ; U Geminorum, 116 ; 
S Cancri, 139 ; variable nebula, 292 

Hipparchus, methods of observation, 
12; cluster in Perseus, 239 ; Andro- 
meda nebula unnoticed by 268 

1 — r 





Holden, colours and magnitades of 
double stars, 156 ; yisibility of star 
in trapezium, 217 ; star in trifid 
nebula, 219; helical nebulie, 258, 
259, 288; annular, 260, 261; 
structure of a collection of stars, 
274 ; stars in Orion nebula, 281 ; 
forms of nebulas, 288; changes in 
Orion nebula, 294 ; irresolvability 
of Milky Way, 368 ; solar transla- 
tion, 386 ; star groups, 893 

Holwarda, periodicity of Mira, 110 

Homann, solar translation, 328 

Homer, constellations, 2 ; catalogue of 
ships, 33 

Hooke, twinkling of stars, 4 ; duplicity 
of 7 Arietis, 163; fifth star in 
trapezium, 217 ; number of Pleiades, 

Hough, quadruple stars, 215 

Houzeau, star-enumeration, 1 ; light- 
aberration of binaries, 202 

Huggins, spectroscopic determination 
of stellar movements, 18 ; gelatine 
plates, 23 ; intrinsic faintness of 
nebulse, 29; photographing stellar 
spectra, 31, 37 ; ultra-violet hy- 
drogen lines, 38-9 ; photograph of 
Sirian spectrum, 47 note ; spectrum 
of Aldebaran, 46 ; of Betelgeux, 54 ; 
identity of chief nebular line, 76 ; 
photographic spectrum of Orion 
nebula, 78-80, 85, -277 ; visual, of 
Andromeda nebula, 81, 105, 272 ; of 
T Goronse, 99 ; of Nova Andromedie, 
105; of components of Cygni, 
161; of cluster in Hercules, 250; 
nebulous star, 253 

Huggins, Mrs., spectroscopic re- 
searches, 76, 78-80, 277 ; star- 
patterns, 394 

Humboldt, first view of 47 Touoani, 

Huygens, P Cygni, 70; 6 Orionis, 163 ; 
Orion nebula, 277 

Hyades, involved in a Milky Way 

outlier, 355 
HydraB, B, light-changes, 114, 125 
Hydrogen, spectrum, 38-9, 69 ; repre- 
sented in stars, 37, 40, 42, 44, 45, 
50, 51, 58, 66, 69, 70, 99, 100, 102 ; 
fluctuating effects, 49, 60, 67; 
invisible in fourth type stars, 63; 
bright in nebulae, 77, 78 

Indi, c, parallax, 309 
Iron, a constitruent of stars, 87, 42, 

Jacob, 61 Cygni, 169; parallax of 

a Heroulis, 307 
Janson, F Cygni, 70, 99 

Eaiseb, colour of Aroturus, 147 
Eazwini, number of Pleiades, 223 
Eeeler, spectrum of y CassiopeiaB, 67 ; 

of Wolf-Bayet stars, 72 
Eempf, stellar photometric review, 

Kepler, new star, 97, 98 ; distances of 

the stars, 297 
Kirch, X Cygni, 114 ; cluster in Libra, 

Kirkwood, origin of double stars, 204 
Klein, colour-change of a UrssB, 147 ; 

form of sidereal system, 395 
Klinkerfues, stellar variability, 122 
Knorre, star-colours, 147 
Knott, maxima of U Geminorum, 115 ; 

duplicity of U Tauri, 183; varia- 
tions of U Cygni, 185 
Konkoly, variable stellar spectra, 49, 

67 ; spectrum of U Orionis, 113 
Kovesligethy, star-colours, 147, 148 

Krliger, stellar parallax, 306, 307 

liACAiLLE, discoveries of nebulae, 10 ; 

observations of southern stars, 335 
Lalande, red stars, 150; observation 

of a nebula, 257 
Lambert, Milky Way, 350 
Lament, Orion trapezium, 277 
Langley, multiple star in trifid nebula, 

Laplace, dark stars, 171 
Lassell, observations of a cluster, 235 ; 

of nebulas, 257, 259, 278 
Law of gravitation in stellar orbits, 

187, 192 
Least squares, principle of, 322-3 
Le Gentil, attendant of Andromeda 

nebula, 270 
Leonis, a. See Begulus 
Leonis, 7, colour change, 158; in- 
trinsic brilliancy, 203, 204 
Leowitz, supposed stellar apparitions, 

Leporis 45, multiple star, 215 
Leporis B, crimson star, 63, 151 
Le Sueur, spectrum of v Arg^, 120 
L'Hermite, total number of stars, 7 
Libra, binary star in, 170; globular 

cluster, 245 
Librae, jS, green star, 150 
— 8, variations in light, 136, 140 

B B 





Lick refnotor, light-oolfecting power, 

Light-curves of variable stars, 112, 
116, 118, 125, 131, 132, 133, 134, 
141, 143 

Light, extinction of, 360, 381 

Light-equation in stellar orbits, 202, 

Light-ratio, 8, 20, 312 

Lindauer, new star in Cassiopeia, 97 

Lindemann, light scale of Duroh- 
musterung, 337 

Liveing and Dewar, energy of electrical 
discharge, 45 

Lockyer, hydrogen-lines in stellar 
spectra, 40 ; dissociation of calcium, 
44; meteoric theory, 55-7, 86-8, 
94, 123, 185; spectrum of hydro- 
gen, 69; of 7 ArgAs, 71; of R 
Oeminorum, 74 ; of nebule, 75 ; 
comets and nebulao, 77, 285 ; deve- 
lopment of stars, 85, 88-9, 103 ; 
origin of binaries, 91, 123 ; colli- 
sion-theory of temporary stars, 
106-7 ; of variables, 123, 127 ; na- 
ture of sunspots, 126 ; Andromeda 
nebula, 272 

Lohse, J. G., observations of Nova 
Cygni, 102 

— O., photographic reticle, 26 
Loomis, period of i) Argills, 118 
Lyncis 12, triple system, 208 
Lyra, ring-nebula in, 260-1 
LyrflB a. See Vega. 

— /5, spectrum, 67-8, 72, 74, 131 ; 
light-changes, 130-1 

— c, quadruple system, 218-4 

Maclxab, observation of ^ Argfia, 119 
Miidler, hypothetical parallax, 195; 

system of ( Scorpii, 208 
Maestlin, Pleiades seen by, 223 
Magellanio Cloud, Greater, area, 274, 
370; nebular contents, 276, 358, 
370; relation of size to distance, 
371; connection with nebular 
system, 372, 876 ; spherical shape, 

Lesser, 47 Touoani near, 246 ; 

nebular contents, 870 ; connection 
with nebular system, 372, 376 
Magnetic influence on scintillation, 6 
Magnitudes, of stars, 3, 20, 22, 312 
Maia, magnitude, 223 ; light-power, 
226; dependent nebula, 231, 232, 
Main, parallax of 61 Cygni, 808 ; 
solar apex, 324 


Mariotte, blue stars, 155 

Marius, detection of Andromeda ne- 
bula, 268 

Mason and Smith, nebular observa- 
tions, 277, 294 

Masses, Of binary stars, 168, 169, 171, 
173, 193, 195 

Maunder, bright lines in stellar 
spectra, 59, 67, 131 

Maurolycus, star in Cassiopeia, 97 

Maury, Miss A. C, spectrum of Pleione, 

Mayer, C, colours of double stars, 155 ; 
detections, 163, 206, 214 

Mayer, T., division of ( Cancri, 209 

Meadows, observations of a Centaori, 

M^hain, discovery of a nebula, 255 

Merope, magnitude, 223; connected 
nebulae, 230, 231 

Messier, nebular discoveries, 10, 235 
note ; cluster in Hercules, 247 ; M 97, 
255; Omega nebula, 282 

Meteoric theory, 55-7, 75, 86-8, 94, 

Michel], red stars, 150 ; number of 
Pleiades, 223 

Micrometer, filar, 15, 304 

Milky Way, a collection of star-groups, 
9, 282, 359-60, 363, 370 ; preference 
for, of red stars, 65, 154, 155 ; of 
gaseous stars and stellar nebulae, 
74, 76, 374 ; of temporary stars, 97, 
106, 369 ; of short period variables, 
144 ; extensions of, towards nebulas 
and clusters, 271, 356, 387 ; groups 
photographed, 273, 356; relation 
to solar apex, 328, 831 ; ecliptic of 
the stars, 850-1 ; visual e£Fect, 352- 
8; bifurcation, 853-4, 385; va- 
cuities, 355-6, 361, 362 ; unresolv- 
able patches, 367-8 ; stratum-theory, 
368-60; annular, 361,378; spiral, 
862 ; intricate organisation, 363-4, 
366; distance, 364-6; significance 
in sidereal system, 368, 369, 396 ; dis- 
tribution of nebulflB and clusters re- 
lative to, 371-4, 376, 378; star- 
counts relative to, 382-3 ; grouping 
of lucid stars, 384 

Mira Ceti, spectrum, 53, 56, 58-9, 69, 
112 ; physical condition, 61; colour, 
111, 152 ; periodicity, 110-12, 124, 
125 ; proper motion, 144, 185, 317 

Mitchel, composite stars, 214 

Monck, mass-brightness of binaries, 
202, 203 ; average stellar parallax, 

Montigny, twinkling of stars, 4, 6 






Moyements, stellar, in line of dghti 18, 

19, 328, 341-8, 391 
Miiller, stellar photometric reyiew, 22 

Nebula, Andbomeda, yisible to the eye, 
10,268; speotrum, 81, 105,272; stel- 
lar outburst in, 104-6, 369 ; structure 
and constitution, 269-72 ; attendants, 
269, 272, 274, 291; development, 
273-4, 396 ; relation to Milky Way, 
356 ; not an external galaxy, 369- 
Nebula, annular, in Lyra, 260-1 
Nebula, Dumb-bell, true form, 266; 
photograph of, 267 ; resemblances 
to, 370 
Nebula, helical planetary, 257-8 
Nebula, Eey-hole, 116, 282 
Nebula, Lagoon, 249, 284, 377 
Nebula, Looped, spectrum, 81 ; situa- 
tion in Magellanic cloud, 276, 374 
Nebula, Omega, 282-3, 294 
Nebula, Orion, observed by Halley, 10 ; 
spectrum, 75, 78-80 ; chemical com- 
position, 81 ; condensed round 6 
Ononis. 216, 277 ; structure, 278-80, 
289 ; nebula minima, 278, 281 ; 
development, 281, 396; variability 
of included stars, 294 ; connection 
with Milky Way, 356, 377, 387 
Nebula, planetary, in Ursa Major, 255- 

Nebula, Trifid, multiple star in, 219, 
283, 295; spectrum, 284; varia- 
tions, 294-5 
Nebulffi, early discoveries, 10; 
numbers, 11 ; intrinsic faintness, 
29 ; spectra, 75-81, 254, 256, 259, 
267, 284 ; analogy with comets, 76, 
262, 285-7, 289, 296; electrical 
condition, 83, 289, 295; supposed 
meteoric constitution, 86-7 ; stellar 
outbursts in, 103-6, 369; stellar 
development from, 204, 219, 250, 
273-4, 281, 296; relations to 
multiple stars, 216-19, 262, 295 ; in 
Pleiades, 230-2, 235, 249, 262, 281, 
290 ; colours, 254, 256, 258, 259 ; 
forms in space, 288 ; immobility, 
290-1, 376, 396; fluctuations of 
light, 291-5; intra-galactic situa- 
tion, 368-73; relations to Milky 
Way, 871-8, 395 
Nebulae, annular, 249, 259-62,289 
NebulaB, cometary, 262, 286-7 
Nebulae, double, 265-6, 290-1 
Nebulae, elliptical, 252, 268, 274-6 
Nebolffi, gaseous, members of a con- 

tinuous series, 84, 368 remoteness, 

257, 374, 375, 378; distribution, 

373-4, 376, 378 
NebuliB, helical, 258, 259, 287-8 
Nebulas, irregular, 252, 268, 276-84 
Nebulae, planetary, spectrum, 77, 254, 

257; occurrence in clusters, 249; 

relationships, 254, 259, 894; com- 
plexities of structure, 255-7, 289; 

no measurable parallax, 257, 317 ; 

satellite stars, 260, 291 
Nebulae, spiral, 263-5, 271, 281; 

possible mode of formation, 287; 

cometary relations, 289 
Nebulae, stellar, monochromatic, 75, 

77, 255 ; limited to Milky Way, 374 
Nebulae, variable in Taurus, 183, 

Nebular hypothesis, 84, 94 
Nebulosities, diffused, 284; in Milky 

Way, 357-8 
Newcomb, approach of a Gygni, 341 ; 

mass of sidereal system, 344, 389. 
New Stars. See Stars, Temporary. 
Newton's law, universal, validity of, 

Niesten, colours of binaries, 160 
NoSl, observation of iy Argfts, 117. 118 ; 

colour of 7 Crucis, 151 
Nova Andromedae, 104-6, 164, 369 
Nova Cygni, 77, 100-3, 105, 106, 164, 

161, 289 ; observations for parallax, 

Nova Ophiuchi, 99 
Nova Serpentarii, 98 

OiiBEBs, period of x Gygni, 114 

Ophiuchi, 70, colours, 159, 170; re- 
volutions, 171 ; mass, 195 ; spec- 
troscopic determinations, 201 ; 
parallax, 307 

Ophiuchi, U, Algol variable, 186; 
light-change, 142 

Orbits, of binary stars, apparent, 
188-9 ; computations, 191-2 ; actual, 
196-8 ; ellipticity, 204 

Orbits, sidereal, 389 

Orion, spectra of stars in, 40, 49, 74, 
93, 369 

Orionis, a. See Betelgeux 

Orionis, jS. See Bigel 

Orionis, 9, spectrum, 79 ; multiple, 
216-8 ; nucleus of nebula, 216, 277 

Orionis, (T, multiple star, 215 

Orionis, U, light-changes, 118 

Orthochromatic plates, 32 

Oxford, photometry, 21 ; photographic 
parallax, 31, 817 




Parallax, annual, measurementB with 
heliometer, 15, 804, 811-12 ; negative, 
105, 257, 301, 814 ; of binaries, 195, 
800; eatiniated, of Alcyone, 228; 
minuteness, 297, 299, 317 ; theory, 
298 ; differential, 801 ; first genuine 
results, 302-8 ; absolute. 805, 811 ; 
probable errors, 307, 309 ; precau- 
tions, 808 ; average, 811, 812, 813, 
314; photographic, 82, 315-^; 
computed and observed, 890 
Parallax, systematic, 882 
Paris, Photographic Congress, 26, 82 ; 
photographs of Pleiades, 231 ; of 
cluster in Auriga, 239 ; of ring- 
nebula, 261 
Pegasi, 85, period and mass, 194, 195 ; 

optical attendant, 218 
Periods of binaries, 194-5 
Pemter, experiments on scintillation, 6 
Perseus, double cluster in, 289-40, 

Personal equation, defined, 14 ; . in 

double-star measures, 189 
Peters, G. A. F., Sirian system, 172 ; 

stellar parallax, 305, 311 
Peters, G. F. W., orbit of 61 Gygni, 

169, 192 
Photographic telescopes, 24-6, 31-2 ; 
irradiation, 27 ; star magnitudes, 
28 ; catalogues, 881 note, 883 
Photographs of nebulie, 32, 261, 263-4, 
267, 269-72, 274-5, 279, 286-7 ; of 
stellar spectra, 88, 40, 42, 57, 58 ; of 
nebular spectra, 79-80; of double 
stars, 190 ; of Pleiades, 223-4, 231-3 ; 
of various clusters, 285, 236, 239, 
246, 249 ; of nebulous stars, 262, 
275 ; in Milky Way, 278, 356 
Photography, star-charting by, 7, 24-7, 
881 note; application to sidereal 
research, 23-9 ; photometric, 27-9 ; 
spectroscopic, 30 : radial movement 
determined by, 31, 828, 842-3, 391 ; 
stellar parallax by, 81, 315-16 ; detec- 
tion of variable stars, 294 ; star- 
gauging by, 882 
Photometer, wedge, 21 ; polarising, 21, 

Photometry, stellar, 19-22; photo- 
graphic, 27-9 ; nebular, 29 
Piazzi, nebulous star, 253 
Pickering, E. G., stellar photometry, 
20 ; sidereal photography, 24-5, 27 
383-4; bright-line spectra, 58, 71, 
75 ; discoveries of stellar nebulae, 
75, 77, 255; classification of vari- 
ables, 96, 120; theories of varia- 
bility, 185 ; changes of Al^l, 


136-7 ; mass-brightness of binaries, 
202, 203, 211 ; close companions of 
( UrsflB and fi Aurigs, 212-3 ; lost 
Pleiad, 222; spectra of Pleiades, 

Pickering, W. H^ photograph of Orion 
nebula, 281 

Pigott, light-changes of B Coron»» 
120 ; of i| Aquils, 133 

Plassmann, stellar variability, 122, 127, 

Plateau, retinal impressions, 4 

Pleiades, a typical cluster, 10, 234; 
spectra, 69, 229; development, 85, 
93, 281, 396 ; traditions concerning, 
220 2; proper motion, 220, 224, 
225; number of stars, 223, 382; 
measurements, 224-5 ; estimated 
distance, 226 ; internal complexity, 
227, 233; nebnlie involved, 230-2, 
289, 290 ; hybrid system, 249, 281, 
369 ; relation to Milky Way, 356, 387 

Pleione, spectrum of, 69, 222 

Plummer, total star-light, 7 ; Tyeho's 
star, 98 ; solar translation, 326 

Pogson, new star, 104 ; B Gephei, 119, 

Pole-star, navigation by, 2; photo- 
metric standard, 21, 22; apparent 
displacements, 298, 299; parallax, 
305, 815; involved in a galactic 
offset, 857 ; radial movement, 391 

Pollux, magnitude, 3 ; radial move- 
ment, 341, 391 ; parallax, 890 

Pond, stellar parallax, 298, 300 ; secular 
aberration, 329 

Position-angles of double stars, 187 

Potsdam, stellar photometry, 22 ; 
spectroscopic researches, 31, 42 ; 
spectrographic determinations, 228, 
842-3, 391 

PrsDsepe, double star in, 239; mea- 
surements, 241; galactic offset 
towards, 356 

Precession of the equinoxes, 13 

Pritchard, stellar photometry, 20-1; 
photographic parallax, 31, 303, 315 ; 
measurement of Pleiades, 225 

Probable errors, 307 

Proctor, star-drift, 846-7; theory of 
Milky Way, 362; distribution of 
nebulflB, 372; star-alignments, 393, 

Procyon, magnitude, 3, 22 ; disturbed 
motion, 173-4; parallax, 814, 890; 
proper motion, 340 ; radial move- 
ment, 843, 391 

Proper motions, defined, 14, 834 ; of 
variables, 144, 346 ; of double stars, 




169, 176, 177, 840; disturbed, 171, 
173, 198 ; of Pleiades, 220, 224, 225 ; 
a criterion of vicimty, 306, 311, 
835-6; parallactio element, 821, 
825, 840 ; residaal, 822, 332 ; abnor- 
mal swiftness, 326, 844-6 ; average, 
336-9 ; relation to magnitude, 839- 
40 ; community in, 346-9 ; relation 
to Milky Way, 860-1 

Ptolemy, constellations, 2; nebulos®, 
10 ; red stars, 146 ; Ononis, 217 ; 
Alcyone ignored by, 222 

Puppis, XJ, yariable double star, 184 

Badau, form of sidereal nniyerse, 895 

Badial component of stellar motion, 

Bambaut, spectroscopic determinations 
of parallax, 200, 201 

Bancken, galactic relations of stellar 
moYements, 851 

Banyard, structure of Orion nebula, 
279, 280 

Bed stars, with banded spectra, 53, 68 ; 
deyelopmental status, 85, 89, 93 ; 
variable in light, 108, 121, 127 ; in 
tint, 152-8; observations, 146-7, 
150-2; varying estimates of inten- 
sity, 148, 151; distribution, 154, 
239, 240 

Beflecting telescopes, advantages for 
photographic purposes, 32 ; effects 
on colour, 149 

Begulus, magnitude, 8 ; blue attendant, 
157, 175 ; parallax, 814, 390 ; radial 
motion, 841 

Bespighi, spectrum of y Argds, 70 

Beticle in photographic star-charting, 

Biocioli, duplicity of i UrssB, 163 

Bichaud, division of a Gentauri, 163 

Bigel, photographed spectrum, 42, 
842 ; variability of visual spectrum, 
49; satellite, 157, 206; recession, 
341-2, 391 

Bight Ascension of stars, 18 

Boberts, successive photographic 
exposures, 24, 294; star-chart, 25 
note ; reflector used by, 32 ; place 
of Tycho*s star, 98; photographs 
of Pleiades, 231 ; of Perseus cluster, 
240 ; of Hercules cluster, 244, 249 ; 
of nebulous star, 253, 275 ; of ring- 
nebula, 261; of spiral, 263; of 
Dumb-bell nebula, 267 ; of Andro 
meda nebula, 269-72 ; of M 81, 274 ; 
of Orion nebula, 279, 281 

Bosse, third Earl, obeervationa ol 


clusters, 285, 288, 248, 246; of 
nebulsB, 261, 253, 260, 262, 264, 267, 
274 ; nebulous star, 253, 254 

Bosse, fourth Earl, double nebula, 

Bussell, measurement of duster about 
IT Grucis, 240 

Rutherfurd, photographs of Pleiades, 
224 ; of PraBsepe, 241 

Sadleb, 70 Ophiuchi and v Cassiopeia, 
171 ; companion of Bigel, 206 

Safarik, colour of U (Jeminorum, 116 ; 
red stars, 121, 149 

SafFord, structure of Orion nebula. 

Sagitts, 10, light-changes, 138; light- 
curve, 148 

Sagittffi, B, periodicity, 181-2 

Savary, orbit of | XJrsce, 192 ; light- 
equation in stellar orbits, 202; 
parallaxes of binary stars, 302 

Sawyer, light-changes of B Goronie, 
121 ; discovery of an Algol-variable, 

Scheiner, stellar spectra, 81, 42, 46, 
68 ; theory of bright-line emission, 

Schiaparelli, Merope nebula, 280 

Schjellerup, observation of U Ophiuchi, 
142; red stars, 160-1; nebulous 
star, 258 

Schjellerup, 152, fourth-type star, 68, 

Schmidt, T Coron®, 99, 100; Nova 
Cygni, 100 ; colours and periods of 
variables, 127; minimum of S 
Cancri, 189 ; red stars, 151 ; T 
Virginis, 183 ; variable nebula, 294 

Schdnfeld, Durchmusterung, 7; T 
GoronaB, 100 ; minima of Mira, 111 ; 
colour of B Cephei, 121 ; observa- 
tions of variables, 139; variable 
nebula, 298; proper motions, 839, 
849 ; related to galactic plane, 350 

Sohroter, Orion trapezium, 277 

Schuster, spectrum of Sohag promi- 
nence, 47 

Schwab, observations of Mira, 112 

Scintillation, of stare, 4-6, 19, 97. 

Scintillometer, 4 

Scorpii, a. See Antarea 

— jS, triple system, 209 

— Vy quadruple star, 214 

— f, triple system, 208 

Searle, nebulous appearance of 

Alcyone, 230 
Secchi, classification of stars, 37, 63 ; 



earbon in foarth-type stan, 63; 
bright-line speotra, 66; red Btare, 
152-154 ; colours of binaries, 159, 
160; satellite of a Heroolis, 183; 
star-dosters, 235, 286; annular 
nebnlsB, 259, 260; star distances, 
317; Milky Way duster, 373; 
star-groupings, 393 

Seeliger, system of CCancri, 210 

Seneca, colour of Sirius, 146 

Seven Stars, 221, 341 

Sidereal Astronomy, 11, 14, 16 

— physics, 16, 66 

' — system, attractive power, 344, 389 ; 
permanence, 345>6, 360 ; nature of 
problem presented by, 379-81; a 
federation, 392; general survey, 
395 ; origin and destiny, 396-7 

Sidereal unit, 387, 388 

Signs of the Zodiac, 2 
'Sirian stars, nature, 87-43, 49-61 ; de- 
velopment, 86-7, 88, 90 ; variability, 
108, 182 

Sirius, magnitude, 3, 22 ; twinkling, 4, 
6; spectrum, 40, 42, 47 note; 
satellite, 92, 172, 191,'^ 199 ; colour, 
146; disturbed motion, 171, 199; 
mass, 172, 173, 195; revolutions, 
173, 198; parallax, 309; proper 
motion, 821, 840 ; radial movement, 

Smyth, Admiral, star-colours, 150, 158 
159 ; observations of clusters, 285 

r— Piazzi, colour-change of 95 
Herculis, 159 

Solar cluster, composition, 885-7; 
dynamical equilibrium in, 392 

-> stars, nature, 44-50 ; development, 
85, 87, 88, 91; variable in short 
periods, 108 

South, colours of double stars, 158 
159 ; catalogue, 170 ; 5 Cygni 181 ; 
CCancri, 209 

Spectra, stellar, type I., 37-42"; type 
II., 44-7 ; type HI., 53-60 ; type 
IV., 62-5 ; variable, 49, 60, 67, 68 
bright lines in, 58-60, 66-72, 79, 86 ; 
of double stars, 92, 161, 178, 184, 
185 ; of Novffl, 99-103, 106, 161 ; of 
variables, 108, 112-4, 116, 122, 126, 
129; of Algol-stars, 144; of 
Pleiades, 229 ; of clusters, 240, 260 ; 
of swift stars, 346 

Spectrograph, 161, 228, 342 

Spectroscopic stellar survey, 30, 58, 

Spectrum, of nebulas, 75-81 
Spectrum analysis, stellar, 17-19| 
80-1, 36 


Spenoer, distribution of nebnlse, 

Spica, spectrum, 87; spectrographio 
discovery of satellite, 213 

Stars, number visible, 1 ; designationa, 
2 ; distances, 2, 14, 297-318, 364-6 ; 
magnitudes, 8, 20, 22, 312; 
twinkling, 4-6 ; total numbers, 6-^, 
382-3; total light, 7, 380; deter- 
mination of places, 13-14; proper 
motions, 14, 320-2, 332-51 ; radial 
movements, 18, 31, 327, 391 ; photo- 
metric measures, 19-22 ; by photo- 
graphy, 27-9 ; distribution, 19, 27, 
858-67, 882-5, 895; photographic 
charting, 26-7, 382-8 ; sunlike na- 
ture, 85-6, 61, 864 ; chemical com- 
position, 87, 42, 44, 46, 54, 56, 62, 
64, 71 ; temperatures, 48, 59, 60, 83. 
108; relative ages, 82, 84-5, 89, 
90-2 ; progress of condensation, 63, 

186, 219 

— groupings of, spectroscopic, 65, 74 ; 
by colour, 164 ; in clusters, 234, 
285-6, 238-9, 246 ; common move- 
ment, 346-9 ; in Milky Way, 364, 
892 ; typical forms, 898-4 

— double, methods of observation, 15, 

187, 189-91 ; origin, 91, 122. 186 ; 
relative masses, 92, 172; colours, 
165-62, 167, 168, 170, 176-7, 184-5 ; 
first discoveries, 163 ; catalogues, 
165, 170 ; profusion, 165 ; systems 
formed by, 166-74, 179; masses, 
166, 168, 169, 171, 178, 193, 195-6, 
199-200 ; obscure components, 171- 
4, 199, 210-1, 218; criteria of 
physical union, 174-7; proper 
motions, 174-6, 177, 198; varia- 
bility in light, 179-86; nature of 
orbits, 187-8, 192, 196-8, 204 ; spec- 
troscopic determination, 199-201; 
light equation, 202, 800; mass- 
brightness, 202-4 ; in dusters, 227, 

— gaseous, show a triple spectrum, 
66 ; spectral fluctuations, 67 ; 
physical nature, 68-9; variable in 
light, 70, 222 ; of Wolf-Eayet type, 
71-2, 74 ; nebular rdationships, 72, 
75, 277 ; collect in groups, 74 ; in 
trapezium, 79, 277; absence of 
parallax, 817 ; of proper motion, 

— multiple, 206, 216, 217-19 ; nebular 
relations, 216-19, 262, 277, 295 

— nebulous, 61, 98, 99, 107, 116, 216, 
252-4, 262 

— quadruple, 209-14 



Stars, temporary, twinkling of, 4, 97, 
98 ; spectra, 77, 99, 100-3, 105, 161 ; 
enomerated, 96-7 ; colours, 97, 98, 
99, 100, 105, 154 ; remoteness, 
106, 817 ; Milky Way objects, 106 ; 
theories, 106>7 

— triple, 176, 206-9, 216 

— variable, by endless gradations, 19, 
; classification, 96, 120, 129; 

nomenclature, 100 ; catalogues, 
108; grouping of periods, 109; 
examples, long periods, 110-6 ; 
irregular, 116-21; theories, 122-8, 
134-5 ; short periods, 129-35, 14a- 
5; Algol-type, 135-44, 182; distri- 
bution, 144-6; connection with 
nebulae, 292-294 

— with banded spectra, two kinds, 53; 
redness, 53, 60, 64, 89, 150 ; 
chemistry, 64, 56, 62 ; constitution, 
56-9; variability in light, 60, 66, 
108, 129, 184-5 ; nebulous, 61 ; dis- 
tribution, 65 ; development, 85, 88- 
90, 92; absence of parallax, 317; 
immobility, 346 

Stone, centre of gravity of a Centauri, 
199; Gape Catalogue, 326; star- 
motions, 327 

Strove, F. G. W., star-colours, 152, 
165-6, 158, 159, 160 ; definition of 
a double star, 164, 176; observa- 
tions of 7 Virginis, 179; of 44 
Bootis, 180 ; of 5 Cygni, 181 ; of a 
Herculis, 183; of 42 Comae, 188; 
of f Cancri, 209; triplicity of c 
Equulei, 208; fifth star in trape- 
zium, 217; companion of Atlas, 
227; stellar parallax, 300, 302; 
theory of Milky Way, 360 

Strove, Ludwig, masses of tj Cassio- 
peia, 199 : solar translation, 324-5, 
327 ; velocity, 328 ; average proper 
motions, 336, 337-8 ; revolutions of 
the stars, 350 

— Otto, catalogue of double stars, 
165; satellite of Procyon, 173; 
variable double stars, 179, 181; 
companion of y Andromedie, 206; 
motions of C Cancri, 210; trape- 
zium, 277; light-changes in ne- 
bulffi, 292, 294 

Sun, stellar diameter and magnitude, 
3, 22 ; nature, 8 ; electrical repul- 
sion in, 60 ; stage of development, 
85, 90-1; collisions with comets, 
107; periodicity, 126-6; isolation 
in space, 317 ; translatory motion, 
318-20, 332-3, 389-90 ; direction, 
321-6, 386; rate, 826-8; nature of 


orbit, 329; plane, 330; centn, tSl, 

388 ; association with «tiber stars, 

332, 349, 386-7 
Sunspot analogy ef stellar variability, 

112, 125-6, 134 
Swift, nebulous star, 253 ; nebolons 

zifig, 262; double nebula, 266; 

Omega nebula, 283 

Taubi, a. See Aldebaran 

Tauri, A, Algol- variable, 136; nature 

of light-change, 140 
Tauri, U, variable double star, 183 
Taylor, spectrum of Orion nebula, 78 ; 

of Andromeda nebula, 105; of x 

Cygni, 114 
Tebbutt, revival of i| Argds, 117 ; 

variable double star, 181 
Tempel, Nova Andromedas, 104 ; dis- 
covery of Merope nebula, 230 ; stare 

in ring-nebula, 260 
Tennyson, colour of Sirius, 147 
Thiele, orbit of Castor, 192 
Thome, colour of v Argils, 117 
Tidal e£Fects in variable stars, 122, 135 
Tisserand, law of force in stellar orbits, 

Toucani, {, swift motion, 309, 326, 

345 ; orbit in sidereal system, 389 
Toueani, 47, globular cluster, 244-5 
Transit-instrmnent, 13 
Trapezium of Orion, 79, 85, 216-8 
Trouvelot, drawing of Andromeda 

nebula, 272 
Tycho Brahe, new star, 97, 154 ; 

e Orionis, 217 ; Pleiades, 222 ; 

Andromeda nebula unnoticed by, 

268 ; stellar parallax, 297, 298 

Ubanometbia Aboxntina, 1 ; Nova, 1 ; 
Oxoniensis, 21 

Ursa Major, elliptical nebula in, 274- 

Urs» Majoris, a, colour-changes, 147 ; 
radial movement, 841 

UrsflB Majoris, 0, spectrum, 87 
planetaxy nebula near, 265 

UrssB Majoris, (, first known double 
star, 163; photographic measure- 
ment, 190 ; slowness of revolutions, 
195 ; multiple system, 212, 347-8 

Ursae Majoris, {, revolutions, 192; 
mass, 193; spectroscopic determi- 
nation, 201 ; inferior limit of 
parallax, 300 

Ussher, scintillation during auror». 




Vboa, designation, 2; magnitude, 8, 
22; brightness, 7; Bpectrum, 37, 
88, 42, 21 ; development, 85, 90 ; 
parallax, 805, 314 ; proper motion, 
821 ; radial motion, 341 
Velocities, stellar, 316, 344-5, 387,889 
YtUarceau, light-aberration of binaries, 

Virginis, a. See Spica 
Virginis, 7, duplicity, 163 ; variability, 
179 ; revolutions, 179, 197 ; intrinsic 
brilliancy, 203 
Virginis, 86, quadruple star, 214 
Virginis, Y, variable double star, 182-^ 
Vogel, H. C, photographic determina- 
tions of radial movement, 81, 328, 
342-3; spectra of stars in Orion, 
40, 59 ; spectrum of Aldebaran, 46 ; 
of Betelgeux, 54 ; of 7 GassiopeiA, 
67 ; of Wolf-Rayet stars, 71, 72 ; of 
B Oeminorum, 74 ; classification of 
stars, 85, 86 ; eclipses of Algol, 137- 
8, 143 note ; close companion of 
Spica, 213; spectra of stars in 
Perseus and Hercules dusters, 240, 
250; observations of nebuls, 256, 
259, 267, 293 
Vogel, H. W., photographic spectrum 
of hydrogen, 88 

Wabd, Nova Andromeds, 104 

Webb, tint of U Gygni, 185 ; struotore 


of dasierB, 245, 246 ; stellar groups, 

Weber, magnetic influence on scintilla- 
tion, 6 • 
Wesley, nebulae in Pleiadefi, 231 ; 

drawings of Andromeda nebula, 269, 

270 ; of Orion nebula, 279 
Wilsing, eclipse of Algol, 139 
. Wilson, revolutions of 61 Cjgni, 169 
Winlock, discoveries of double stars, 

175, 194 
Winnecke, Tariability of 8 Mono- 

oerotis, 182 ; observations of 

PrsBsepe, 241 ; variable nebula, 293; 

stellar parallax, 306 
Wolf, C, catalogue of stars in Pleiades, 

223, 2^5; centrifugal tendency 

among, 228; measurements of 

Prtesepe, 241 
Wolf, Max, Nova Andromedas, 104 
Wolf, R., period of t? Argtls, 118 ; 

stellar and solar periodicity, 125 
Wolf-Rayet stars, 71, 72, 74, 317 
Wright, disc- theory of Milky Way, 358 
Wrublewsky, period of 9 Equulei, 194 

TouNo, observations of Nova Andro- 
medjEB, 105 

Zodiac, signs of the, 2 
Z511ner, polarising photometer, 21, 
22 ; age of stars, 85 


sPoroBWOODa and co^ kxw-stbbt squasb 


39 Paternoster Row, London, E.C« 
September 1890. 

©afaloguc of "^ooifes 



English Church in the Eigh- 
teenth Century. By Charles J. 
Abbey and John H. Overton. Cr. 8vo. 

ABBOTT {T. ^.)-Works by. 
The Elements of Logic, 
Elementary Theory of theTides: 

the Fundamental Theories demonstrated 
without Mathematics. Crown 8vo. 2f. 

ACTON.— VioA^xtL Cookery for 
Private Families. By Eliza 

Acton. With 150 Woodcuts. Fcp. 8vo. 
4r. fid. 

A. K. JT.B.—Tht Essays and Con- 
tributions of. Cr. 8vo. 
Autumn Holidays of a Country Parson, 3^. M, 
Changed Aspects of Unchanged Truths, 

Commonplace Philosopher, 3^. 6d. 
Counsel and Comfort from a City Pulpit, 

3J. 6d. 
Critical Essays of a Country Parsoni 3^. 6d. 
East Coast Days and Memories. 3^. 6d, 
Graver Thoughts of a Country Parson. 

Three Series, $s. 6d. each. 
Landscapes, Churches, and Moralities, 3^. 6d. 
Leisure Hours in Town, 3/. 6d. 
Lessons of Middle Age, 3^. 6d, 
Our Little Life. Two Series, 3^. 6d. each. 
Our Homely Comedy and Tragedy, y. ed. 
Present Day Thoughts, 3^, 6d. 
Recreations of a Country Parson. Three 

Series, 31. 6d. each. 
Seaside Musings, 31. 6d, 
Sunday Afternoons in the Parish Church of 

a Scottish University City, 3J. 6d, 
•To Meet the Day' through the Christian 

Year : being a Text of Scripture, with an 

Original Meditation and a Short Selection 

in Verse for Every Day, 41. 6ti. 

ALLEN.— Force and Energy: a 

Theory of Dynamics. By Grant Allen. 
8vo. 7j. 6d. 

American Whist, Illustrated : 

containing the Laws and Principles of 
the Game, the Analysis of the New 
Play and American Leads, and a Series 
of Hands in Diagram, and combining 
Whist Universal and American Whist. 
By G. W. P. Fcp. 8vo. 6s. td. 

.JIMDS.^K Primer of the English 

By Sheldon Amos. Crown 8vo. 6s. 
ANSTE K— TheBlack Poodle,and 

other Stories. By F. Anstey, Author 
of 'Vice Versa.' Cr.Svo. 2j. bds.; 2s. 6d. cl. 

ARISTOTLE.— The Works of. 

The Politics, G. Bekker's Greek 
Text of Books I. III. IV. (VIL) with 
an English Translation by W. £. 
BoLLAND, M.A. ; and short Introductory 
Essays by A. Lang, M. A. Ct.Z^o,^s.6d. 

The Politics : Introductory. Essays. 
By Andrew Lang. (From Bolland and 
limg's.' Politics. Crown 8vo. 2s. 6d. 

The Ethics ; Greek Text, illustrated 
with Essays and Notes. By Sir Alexan- 
der Grant, Bart. M. A. LL.D. a vols. 
8vo. Z2S. 

The Nicomachean Ethics, Newly 

Translated into English. By Robert 
Williams, Barrister-at-Law. Crown 
8vo. 7x. 6d» 

ARMSTRONG (G. i^)— Works by. 

Poems: Lyrical and Dramatic Fcp. 
8vo. 6s. 

King Saul. (The Tragedy of Israel, 

Part I.) Fcp. 8vo. Ss, 

King David. (The Tragedy of Israel, 

Part II.) Fcp. 8vo. 6x. 

King Solomon. (The Tragedy of 
Israel, Part III.) Fcp. 8va 6s. 

Ugone: A Tragedy. Fcp. 8vo. 6s. 
[Continuid on tuxtpagtn 

CATAtoGva or GaidutAL Aim Scaimnc Boob 

ARMSTRONG {G. -K)— Worics by 

— continued, 

A Garland from Greece; Poems. 

Fq>. 8vo. 9/. 

Stories of Wicklow; Poems. Fqx 

8vo. 9x. 

Victoria Regina et Itnperatriz : 

a Jnhtlee Song firom IreUnd, 1887. 410. 

Mephistophdes in Broadcloth : 

a Satire. Fcp. Svo. 4/. 

The Life and Letters of Edmund 
J. Armstrong. Fcp. 8vo. 7/. 6/. 

ARMSTRONG (^./.)- Works by. 
Poetical Works. Fcp. 8va 51. 
EssajTS and Sketches. Fcp. 8to. 5^ . 

ARNOLD {Dr. 7!)— Works by. 

Introductory Lectures on Mo- 
dem History. 8vo. 7^. td. 

Sermons Preadhed mostly in 
the Chapel of Rugby School. 

6 ▼olt.ciown 8vo. 30;. or sepaiately, 5/. ea. 

Miscellaneous Works. 8va7J.6i£ 
ARNOTT.—'BXtxntxA& of Physics 

By Nbzl Arnott, M.D. 8vo. ixr. M 

^£^^K— Notes on Physiology 
for the use of Students pre- 
paring for Examination. With 

1 34 Woodcuts, ByHKN&YAsHBY,M.D. 
Fqp. 8vo. 5i. 

ASHBY and WRIGHT, — The 
Diseases of Children. By henry 

AsHBY, M.D. and Gko.A. W&igut,B.A. 
F.R.C.S. 8vo. au. 

Atelier (The) du Lysj or, an Art 

Student in the Reign of Terror. By the 
Author of ' Mademoiselle Mori.' Crown 
8vo. ax. 6^. 

^^COA:— TheWorksand Life of. 
Complete Works. Edited by 
R. L. Ellis, J. Spedding, and D. D. 
Hbath. 7 vols. 8vo. £Z' 13^* ^ 

Letters and Life, including: all 
his Occasional works. Edited 

I7 J. Spkdding. 7 vols. 8vo. £\, 4/. 

The Essays; with Annotations. By 
Richard WHATKLT, D.D., 8vo. icv. (id. 

The Essays; with Introduction, 
Notes, and Index. By E. A. Abbott, 
D.D. 2 vols. fcp. 8vo. price 6r. Text 
and Index only, without Introduction 
and Notes, in i vol. fcp. 8vo. 2x. ^ 


edited by the Dukb op Bkauiokt.ICG., 
anisted by Alfucd £. T. Watson. 

Huntiiifi^. By the DukA of Beau- 
fort, ICG. and MowBRAT MoRUS. 
With S3 IIlus. by J. StniseM, T. Charlton, 
and A. M. Biddulph, Crown ovo. lOf. 61/. 

Fishing:. By H. Cholmondeley- 

VoL L Salmon, Trout, and Grayling. 

With 158 lUuitiations. Cr. 8vo. lor. fid. 
Vol. II. Pike and other Coarse' Fish. 

With 132 Ilhstrationa. Cr. 8vo. lOf. (id. 

Racing: and Steeplechasing:. ^y 

the Earl of Suffolk and Berkshire, 
W. G. Craven, &c, With 56 Illnstra- 
tioos by J. Stai]gesa. Cr. 8va. lOr. 6dL 

Shooting. By Lord Walsingham 

and Sir Ralph Paynb-Gallwky, Bart. 

VoL I. Field and Covert. With lOS 

Illostratioos. Cr. 8va lOf. 6dL 
Vol. VL Moor and Manh. With 65 lUns- 
trations. Cr. 8vo. lor. td. 

Cycling:* By Viscount Bury, 
K:.C.M.G. and G. Lacy HiLUKR. With 
19 Plates and 61 Woodcnti by Vlsooant 
Bury and Joseph PenneU. Cr. 8vo. lor. 6^ 

AtUetics and FootbalL By 

Montague Shearican. With 6 fnll* 
page Illostrations and 45 Woodcuts by 
Stanley Berkeley, and frcMU Photographs 
by G. Mitchell Cr. 8vo. lOr. 6A 

Boating. By W. B. Woodgate. 
With 10 full-page Illustrations and 39 
Woodcuts in the Text. Cr. 8vo. lOf. 6</. 

Cricket By A. G. Steel and the 
Hon. R. H. Lyttelton. With 1 1 full- 
raige Illustrations and 52 Woodcuts in the 
Text, by Lucien Davis. Cr. 8va lor. dd. 

Driving:. BytheDuKEorBEAijFORT. 

With II Plates and 54 Woodcuts by J. 
Stuiigess and G. D. Giles. Cr. 8vo. lor. dd. 

Fencing,Bozing:, and Wrestling. 

By Walter H. Pollock, F. C. Grove, 
C. Prevost, £. B. MiCHELL, and 
Walter Armstrong With 18 Plates 
and 24 Woodcuts. Crown 8va lOf. dd. 

Golf. By Horace Hutchinson, the 
Rt. Hon. A. J. Balfour, M.P. Andrew 
Lang, Sir W. G. Simpson, Bart. &c. 
With ig Plates and 69 Woodcuts. 
Crown 8vo. lOf. bd. 

Tennis, Lawn Tennis, Rackets, 
and Fives. By j. m. and c g. 

Heathcote, E.O. Pleydell-Bouverie, 
and A. C. Ainger. With 12 Plates and 
67 Woodcuts, &c. Crown 8vo. lOr. (^ 


BAGEHOT {miier)— Works by. 

Biographical Studies. 8vo. 12s. 

Economic Studies. 8vo» loj. 6d. 

Literary Studies. 2 vols. 8vo. sSj. 

The Postulates of English Po- 
litical Economy. Cr. 8vo. :ts» 6d. 

A Practical Plan for Assimi- 
lating: the English and Ameri- 
can Money as a Step towards 
a Universal Money. Ct.SY0.2sM. 

^ JPAGWELJL—lTtlasx^. under the 

TudOrSy with a SucciDCt Account of 
the Earlier History. By Richard Bag* 
WELL, M. A. (3 vols.) Vols. I. and II. 
From the first invasion of the Northmen 
to the year 1578. 8vo. 32r. Vol. III. 
1 578-1603. i&r. 

BAIN {Alexander)— Works by. 
Mental and Moral Science. 

Crown 8yo. lor. 6</. 

Senses and the Intellect^. 
Emotions and the WilL Svais^. 
LoriCy Deductive and Inductive. 

FART I. Deduction^ 41. Pa&t II. In- 
ductMH^ 6s. 6d. 

Practical Essays. Cr. 8vo. 2^. 
BAKER.— B7 the Western Sea : 

a Summer Idvll. By James Baker, 
F.R.G.S. Author of 'John Westacott.* 
Cr. 8va dr. 

BAKER — 'V/ar with Crime': 

being a Selection of Reprinted Papers 
on Crime, Prison Discipline, &c By 
T. Barwick Lu Baker. 8vo. 12s. 6</. 

BAKER (Sir S. ^)— Works by. 
Eight Years in Ceylon. Crown 

8vo. Woodcuts, y. id. 

The Rifle and the Hound in 

Ceylon. Cr. 8vo. Woodcuts, y. 6d. 

^^Z^. — Handbook for Steam 

UserSy being Notes on Steam Engine 
and Boiler Management and Steam 
Boiler Explosions. By M. Powis Bale, 
M.LM.E. A.M.I.C.E. Fcp. 8to. 2s. 6d. 

BALL {TkeRt. Han. J. 7!)— Works 

The ReformedChurch of Ireland 

(1537-1886). 8vo. ^s. dd. 

Historical Review of the Legfis- 
lative Systems Operative in 

Ireland) from the invasion of Henry the 
Second to the Union (i 172-1 800). 8vo. 6j. 

BARKER.— k Short Manual of 
Surgical Operations. ByA.E.j. 

Barker, F.R.C.S. With 61 Woodcuts. 
Crown 8vo. \2s. td, 

BEACONSFIELD (The Earl of)— 
Works by. 

Novels and 

enden Edition. 

Vignettes. 11 
Tancred. SybiL 

Tales. The Hugh- 

With 2 Portraits and il 
vols. Crown 8vo, 42X. 
Henrietta Temple. 
Contarini Fleming, &c 
Alroy, Ixion, &c. 
The Young Duke, &c 
Vivian Grey. 

Novels and Tales. Cheap Edition, 
complete in il vols. Crown 8vo. \s. 
each, boards ; i /. 6d. each, doth. 

BECKER {Prq/essor)—V/orks by. 

Gallus; or, Roman Scenes in the 
Time of Augustus. Post 8vo. 7j. 6di 

Charides ; or, Illustrations of the 
Private Life of the Ancient Greeks. 
Post 8vo. 7i. 6d. 

BELL.-W\VL o' the Wisp : a Story. 
By Mrs. Hugh Belu Illustrated by 
£. L. Shute. Crown 8vo. 31. 6d. 

Handbook of Cryptogamic 

Botany. By a. W. Bennett, M.A. 
B.Sc. F.L.S. and George R. Milne 
Murray, F. L. S. With 378 Illustrations. 
8vo. i6j. 

^^^-AT^rz:— Clinical Lectures 
onVaricoseVeins of theLower 

Extremities. By William H.Ben. 
NETT, F.R.C.S. With 3 Plates. 8vo. 6jr. 

BENTLEY.—K Text-Book of 
Organic Materia Medica. By 

Robt.Bentlby,F.L..S. 6a Illustrations. 
Crown 8to. Is. td. 



BINET.-^Tht Psychic Life of 
Micro-organisms, a stndy in 

Experimental Psychology. By Alfrsd 
BxNET. Crown 8vo. jj. 

BLAKE.— Tzbles for the Conver- 
sion of 5 per Cent Interest 
from tV to 7 per Cent By j. 

Blake, of the London Joint Stock Bank, 
Limited. 8vo. I2s. 6J 

Book (The) of Wedding Days. 

Ammged on the Plan of a Birthday Book. 
With q6 Illustrated Borders, Frontispiece, 
and Title-page by Walter Crane ; and 
Quotations for each Day. Compiled and 
Arranged by K. £. J. Reid, May Ross, 
and Mabel Bam field. 4to. 2ix. 

BOULTBEE.—Commtntairj on 
the 39 Articles of the church of 
England. By the Rev. T. P. Boultbee. 
Crown 8vo. os, 

BOURNE 0^«)-Works by. 
Catechism of the Steam Engine 

in its various Applications in the Arts ; to 
which is now added a chapter on Air and 
Gas Engines, and another devoted to 
Useful Rules, Tables, and MemoTanda, 
212 Woodcuts. Ctown 8vo. 7x. td^ 

Handbook of the Steam Engine ; 

a Key to the Catechism of the Steam 
Engine. With 67 Woodcuts. Fcp.8vo. 9/. 

Recent Improvements in the 

Steam Engine. With 124 Woodcuts. 
Fcp. 8vo. 6x. 

^OWTiV:— Thirty Years of Co- 
lonial Government: a Selection 
from the Official Papers of the Rt. Hon. 
Sir George Ferguson Bowen, G.C.M.G. 
D. C. L. LL. D. ^c Edited by Stanley 
Lane-Poole. 2 vols. 8vo. 32J. 

BRASSEY(Lady)—yNoxk!^ by. 

A Voyage in the 'Sunbeam,' our 
Home on the Ocean for 
Eleven Months. 

Library Edition. With 8 Maps and 
Charts, and 1 1 8 Illustrations, 8vo. 21/. 

Cabinet Edition. With Map and 66 
IllustratioDs, crown 8vo. *js. 6</. 

School Edition. With 37 Illustrations, 
fcp. 2j. cloth, or 3^. white parchment. 

Popular Edition . With 60 Illustrations, 

. 4to. td. sewed, u. cloth. 

BRASSEY {Lady) —Works by 


Sunshine and Storm in the East 

Library Edition. With 2 Biapt and 
114 lUustxations, 8vo. 21/. 

Cabinet Edition. With 2 Maps and 
- 114 Illustrations, crown 8vo. 71. 6/. 

Popular Edition. With 103 lUnstia- 
tions, 4ta 6^ sewed, ix. doth. 

In the Trades, the Tropics, and 
the ' Roaring Forties.' 

Cabinet Edition. With Map and 220 
Ulostxatiaos, crown 8vo. 7x. &£ 

Popular Edition. With 183 Illustra- 
tioQs, 4to. 6^ sewed, u. doth. 

The Last Voyu'e to India and 
Australia in the ' Sunbeam.' 

With Charts and Maps, and 40 Illustrations 
in Monotone (20 full-page), and nearly 200 
Illustrations in the Text from Drawii^ 
by R. T. Pritchbit. 8vo. 2Ij. 

Three Voyages in the 'Sun- 
beam.' Popular Edition. With 346 
Illustrations, 4to. 2x. dd, 

BRAY.—TYat PhUosophy of Ne- 

CeSSity ; or, Law in Mind as in Matter. 
By Charles Bray. Crown 8vo. 5^. 

Brinkley's Astronomy. Re-edited 

and Revised by J. W. Stubbs, D.D. 
and T. Brunnow, Ph.D. With 49 
Diagrams. Crown 8vo. 6f. 

BROWNE.— AxL Exposition of 

the 39 Articles, Historical and 
Doctrinal. By E. H. Browne, D.D., 
Bishop of Winchester. 8vo. i6r. 

BRYDEN^KLoof and Karroo: 

sport, Legend, and Natural History in 
Cape Colony. By H. A. Bryden. With 
17 illostrations. 8va lar. 6d. 

BUCKLE.-Historj of Civilisa- 
tion in England and France, 
Spain and Scotland. ByHENRY 

Thomas Buckle. 3 vols. cr. 8vo. 24s. 

Food and Home Cookery. With 

II Woodcuts. Crown 8vo. 2x. 6d. 

Health in the House. With 41 

Woodcuts and Diagrams. Crown 8vo. 2x. 


BULL (Thomasy-^OT'k& by. 

Hints to Mothers on the 
Management of their Health 

during the Period of Pregnancy. Fcp. 8vo. 

The Maternal Management of 
Children in Health and Dis- 
ease. Fcp. 8yo. IS, 6d. 

BULLINGER.—h Critical Lexi- 
con and Concordance to the 
English and Greek New 
Testament. By e. w. bulunger, 

D.D. Royal 8vo. 1$/. 

BURL>ETT,— Y^nncty Princess, 

and People ; anAccountof the Social 
Prepress and Development of our own 
Times, as illustrated by the Public Life and 
Work of their Royal Highnesses the Prince 
and Princess of Wales, 1863-1889. By 
Henry C. Burdett. 8vo. 21s, 

BUTLER (&w«^/)— Works by. 
• Op, I. Erewhon. Cr. Svo. 55. 
Op. 2. The Fair Haven. A Work 

in Defence of the Miraculous Element in 
our Lord's Ministry. Cr. Svo. 7x. td. 

Op. 3. Life and Habit An Essay 

after a Completer View of Evolution. 
Cr. 8vo. 7x. td. 

Op. 4. Evolution, Old and New. 

Cr. Svo. loj. 6d, 

Op. S Unconscious Memory. Cr. 

Svo. ^s, td. 

Op. 6. Alps and Sanctuaries of 
Piedmont and the Canton 

TicinO. illustrated. Pott 4to. lOf. 6^. 

Op.7. Selections from Ops. i-6. 

With Remarks on Mr. G. J. Romanes' 
< Mental Evolution in Animals.' Cr. Svo. 
is. 6d, 

Op. 8. Luck, or Cunning, as the 
Main Means of Organic 
Modification ? Cr. Svo. 71. ed. 

Op. 9. ExVoto. An Account of the 
Sacro Monte or New Jerusalem at Varallo- 
Sesia. los, 6d. 

Holbein's 'La Danse.' A Note on 

a Drawing called 'La Danse.' y, 

CARL YLE. — Thomas Carlyle: 

a History of his Life. By J. A. Froude. 
Vols. I. and IL 1 795-1 835, crown Svo. 
7J. Vols. III. and IV. 1834^1881, 
crown 8vo. 7/. 

CARUS. — Fundamental Prob- 
lems. The Method of Philosophy as a 
Systematic Arrangement of Knowledge. 
By Dr. Paul Car us. Crown Svo. 4f. 

CASE,—'9\xy%vcdX Realism: being 

an Analytical Philosophy from the Physical 
Objects of Science to the Physical Data 
of Sense. By Thomas Case, M.A. 
Fellow and Senior Tutor C.C.C. 8vo. 15X. 

CHILD, — Church and State 

under the Tudors. By Gilbert 
W. Child, M.A. Exeter College, Oxford. 
8vo. 15J. 

CHISHOLM. — Handbook of 
Commercial Geography. By 

G. G. Chisholm, B.Sc. With 29 Maps. 
8vo. 16^. 

CHURCHILL. — Speeches by 

Selected and Edited, with an Intrpduction 
and Notes, by L. J. Jennings. 2 vols. 
Svo. 24r. 

ClaverSi the Despot's Champion: 

a Scou Biography. By A Southern. 
Crown Svo. ys, 6d. 

CLERK.— Th^ Gas Engine: By 

DUGALD Clerk. With loi Illustrations 
and Diagrams. Crown 8vo. *js. 6d. 

CLODD.—Tht Story of Creation : 

a Plain Account ot Evolution. By 
Edward Clodd. With 77 Illustrations. 
Crown 8vo. p. 6d. 

in Arctic Seas. By W. J. Clutter- 
BUCK, one of the Authors of * Three in 
Norway.' With 39 Illustrations. Cr. 8vo. 
lox. 6d, 

COATES.—A Manual of Path- 
Ology. By Joseph Coats, M.D. 
Pathologist to the Western Infirmary, 
Glasgow. With 364 Illustrations. 8vo. 
3»j. ^. 

COLEJVSO.-^ThePentattnch and 
Book of Joshua Critically Ex- 
amined. By J. W. COLENSO, D.D. 
late Bishop of Natal. Crown 8vo. 6s, 

COMYNi—Atherstone Priory: a 

Tale. ByL,N. COMYN. Cr.Syo.2s.6d, 

CONDER.^h Handbook to the 

Bible, or Guide to the Study of the Holy 
Scriptures derived from Ancient Monu- 
ments and Modem Exploration. By F. 
R. CONDER and C. R. Conder. Crown 
8vo. is. 6d, 

CATALOGxm ow GMNMMdt AND Scnumnc Boors 

The ^neid of Virril. Thms- 

lated into English Verse. Crown Sva 6s, 

The Poems of VireiL Translated 
into Englisk Prose, Crown 8vo. 6f. 

The Life and Epistles of St 

PauL By the Rev. W. J. Contbbakb, 
M. A. and the Veiy Ker. J. S. HowsoN, 
libnury Edition, a ifols. 8to. %u. 
Student's Edition. I toL crown 8vo. 6r« 

CC7(9ir^.— Tablets of Anatomy. 

Bj Thomas Cooke, F.R.C.S. Eng. 
B. A. B.Sc. M.D. l^aris. Post 4to. V* 6d. 

COOZ/DGE.Smss Travel and 
SwissGuide-Books. By w. a. b. 

CoouDGB, Fellow of Magdalen College, 
Oxford. Crown 8vo. lo/. 6d, 

CEEIGIfTOI\r.— History of the 
Papacy During the Reforma- 
tion. BytheReT.M.CuuoBTON,M.A. 
8vo. Vols. I. and IL 1 378-1464, 32/. ; 
Vols. III. and IV. 1464-1518, 34^. 

CROOJCES.Stl^ct Methods in 
Chemical Analysis (chiefly inor- 
ganic). By William Crookks, F.R.S. 
V.P.C.S. With 37 lUttstrations. 8m 241. 

CRUMP (^.)-Works by. 
A Short Enquiry into the Form- 
ation of Political Opinion, 

from the Reign of the Great Families to 
the Advent of Democracy. 8va 7/. 6tL 

An Investigation into the Causes 
of the Great Fall in Prices 

which took place coincidently with the 
Demonetisation of Silver by Germany. 
8vo. dr. 

COZZ-ffK— Handbook of Prac- 
tical Telegraphy. By R. s. 

CULLXY, VL Inst. C.E. 8vo. i6f. 

CW?z<9-A?: — Russia in Central 
Asia in 1889 and the Anglo- 
Russian Question. B/the Hon. 
George N. Curzon, M.P. 8vo. 21/. 

DA VIDSON ( W. Z.)— Works by. 

The Logic of Definition Ex- 
plained and Applied. Cr.8vo.6f. 

Leading and Important English 
Words Explained and Ex- 
emplified, t'cp. 8vo. 3J. &/. 

DE REDCLIFFE.—The^ Life of 
the Riffht Hon. Stratford Can- 
ning : Viscount Stratford De 

Redclifie. By Stanley Lanb-Poole. 
With 3 Portraits, a vols. 8vo. 36ir. 

DE SALIS (Jtfrr.)— Works by. 
Savouries k la Mode. Fcp. Sva 

I/, hoards. 

Entrtes k la Mode. Fqx 8vo. 

IX. 6d, hoards. 

Soups and Dressed Fish k la 

mode. Fcp. 8vo. It. 6d. hoards. 

Oysters k la Mode. Fq>. 8va 

I/. 6d, boards. 

Sweets and Supper Dishes k la 

Mode. Fcp. 8vo. u. 6d, boards. 

Dressed Vegetables k la Mode. 

Fcp. Svo. is. od. boards. 

Dressed Game and Poultry k 

la Mode. Fcp. 8va IS. 6tL boards. 

Puddings and Pastry k la Mode. 

Fcp. 8vo. is. 6d, boards. 

Cakes and Confections k la 

Mode. Fcp. 8vo. ij. 6^. bocurds. 

Wrinkles and Notions for every 

Household. Crown 8vo. 2s. 6d. 

Tempting Dishes for Small 

Incomes. Fcp. 8va is. 6d. 

CmCf in America. By Alexis ds 
TocgaEViLLB. Transkted by Hrnry 
Reevb» CB. 2 vols, crown 8vo. i6x. 

DELAND (ifrx.)— Works by. 
John Ward, Preacher: a Story. 

Crown Svo. Cabinet Edition, 6f. ; Popu- 
lar Edition, ax. boards, ax. 6d. doth. 

The Old Garden, and other Verses. 
Fcp. Svo. 5x. 

Florida Days. With la FuU-page 
Plates (2 Etched and 4 in Coloors), aod 
about 50 lUastratioDS in the Text, by 
Louis K. Harlow. Svo. au. 

DICKINSON{ W. ^)— Works by. 
On Renal and Urinary Aflfec- 

tions. With 12 Plates and laa Wood- 
cuts. 3 vols. Svo. £^. 4s. 6tL 

The Tongue as an Indicator of 

Disease : being the Lumbeian Lectures, 
delivered at the Royal College of 
Physicians in March iS&. Sva yj. 6d. 


JDOWEZZ.'^A History of Taxa- 
tion and Taxes in Eng^land 

from the Earliest Times to the Year 1885. 
By Stephen Dowell. (4 vols. 8vo.) 
Vols. I. and II. The History of Taxation* 
2is, Vols. III. and IV. The History of 
Taxes, si/. 

DO YLE (/. ^.)-Works by. 

The English in America: Vir- 

aia, Maryland, and the 
rolinas. 8vo. i8x. 

The English in America: The 
Puritan Colonies. 2 vols. 8vo. 

DOYLE (A. 0;«fl«)— Works by. 
Micah Clarke: his Sutement as 

made to his three Grandchildren, Joseph, 
Gervas, and Reuben, during the hard 
Winter of 1734. With Frontispiece and 
Vignette. Crown 8vo. 3/. 6^. 

The Captain of the Polestar; 

and other Tales. Crown 8vo. 6i. 

Dublin University Press Series 

(The) : a Series of Works undertaken 

by the Provost and Senior Fellows of 

Trinity College, Dublin. 
Abbott's (T. K.) Codex Rescriptus Dublin- 

ensis of St Matthew. 4to. 21s, 
■ Evangeliorum Versio Ante^ 

hieronymianaex CodiceUsseriano (Dubiin- 

ensi). 2 vols, crown 8vo. 3i^. 
Allman's (G. J.) Greek Geometry from 

Thales to Euclid. 8vo. los, 6d, 
Bumside (W. S.) and Panton's (A. W.) 

Theory of Equations. 8vo. 12/. 6^. 
Casey's (John) Sequel to Euclid's Elements. 

Crown 8vo. y. 6d, 
— «._ Analytical Geometry of thp 
' Conic Sections. Crown Svo. 7^. 6d, 
Davies' Q. F.) Eumenides of iEschylus. 

With Metrical English Translation. 8vo. 

Dublin Translations into Greek and Latin 

Verse. Edited by R. Y. Tyrrell. 8vo. 

12s. 6d, 
Graves' (R. P.) life of Sir William 

Hamilton. 3 vols. 15J. each. 
Griffin (R. W.) on Parabola, Ellipse, and 
. Hyperbola, Crown 8vo. is. 
Hobart's (W. K.) Medical Language of St. 

Luke. Svo. i6r. ' 
Leslie's (T. E. Cliffe) Essays in Political 

Economy, 8va lox. 6d, 
Macalister's (A.) Zoology and Morphology 

of Vertebrata. 8vo. lOr. 6d, 
BiacCullagh's (James) Mathematical and 

other TracU. Svo. 15X. 

Dublin University Press Series 

(The) — tontinued, 
Magttire's (T.) Parmenides of Plato, Text 

with Introduction, Analysis, &C. Svo. 71.6^/. 
Roberts' (R. A.) Examples in ihe Analytic 

Geometry of Plane Curves. Fcp. Svo. 5^. 
Sonthey's(R.} Correspondence with Caroline 

Bowles. Edited by E.Dowden. Svo. 141. . 
Stubbs* (J. W.) History of the University 

of Dublin, from its Foundation to the End 

of the Eighteenth Century. Svo. izr. 6c/. 
ThomhiU'? (W. J.) The iEneid of Virgil, 

freely translated into English Blank 

Verse. Crown 8vo. 7/. t(L 
Tyrrell's (R. Y.) Cicero's Correspondence. 

Vols. I. II. and IIL Svo. each I2J. 
The Achamians of Aristo« 

phanesy translated into English Verse. 

Crown Svo. 2j. 6</. 
Webb's (T. E.) Goethe's Faust, Transla- 
tion and Notes. Svo. \2s, dcU 
The Veil of Isis : a Series 

of Essays on Idealism. Svo. lor. dd. 
Wilkins' (G.) The Growth of the Homerio 

Poems. Svo. 6j. 

EAEZ,— The Elements of Labo- 
ratory Wort A Course of Natural 
Science for Schools. By A. G. Earl. 
Crown Svo. 41. 6^. . 

EDERSHEIM {Dr. A. E.) — 

Works by. 
The Life and Times of Jesus 
the Messiah. 

Library Edition, 2 vols. Svo. 24J. 

Jesus the Messiah: being an 

Abridged Edition of * The Life and Times 
ofjesus the Messiah.* i vol. cr. Svo. 71. 6^. 

Propheqr and • History in ref- 
lation to the Messiah : the War- 

burton Lectures, 1880.1884. Svo. I2x. 

ELLICOTT (C. /, D.D. Bishop of 
GlaucesUr and J5risfoi)--Works 

A Critical and Grammatical. 
Commentary on St Paul's 
Epistles. Svo. 

L Corinthians. i6x. 
Galatians. Ss.6d. 
Ephesians. &r. 6d. 
Pastoral Epistles. lOf. &/. 
PhilippianSi Colossians, and PhilemofL 

lOf. 6d. 
Thessalonians. Js, 6d. 

Historical Lectures on the Life 
of Our Lord* Jesus Christ. 




8vo. is, each, sewed ; is, 6d. each, doCA. 

Darwin. By Grant Allen. 

MarlborouglL By G. Saintsbury. 

Shaftesbury (7^ JFi'rst Earl), By 
H. I). Traill, 

Admiral Blake. By David Hannav. 

Raleigh. By Edmund Gossf. 

Steele. By Austin Dobson. 

Ben Jonson. By J. A. Symonds. 

Canning. By Frank R Hilu 

Claverhouse. By Mowbray Morris. 

ERICHSEN {John wffw)— Works 

The Science and Art of Surgery: 

Being a Treatise on Surgical Injuries^ 
Diseases, and Operations. With 1,025 
Illustrations. 2 vols. 8vo. 48/. 

On Concussion of the Spine, 

Nervous Shocks, and other Obecore In- 
juries of the Nervous System. Cr. Svo. 
lor. 6</. 

EWALD {ffeinnchy-^Vfoxkshy. 

"The Antiquities of Israel. Trans- 
lated from the German by U. S. SOLLY, 
M.A. Svo. 12J. 6d. 

The History of Israel Trans- 
lated from the German. 8 vols. 8vo. 
Vols. I. and XL 24^. Vols. IIL and 
IV, 21/. Vol. V. i8f. VoL VI. i6j. 
VoL VII. 2fj. VoL VIII. with Index 
to the Complete Work. i8x. 

FARRAR.^l^nf^zZ^ and Lan- 
guages. A Revised Edition of Chap- 
ters OH Language and Families 0/ Speech, 
By F. W. FARkAR, D.D. Crown 8vo. 6x. 

-F/^r-^-- Nation Making: a Story 

of New Zealand Savageism and Civil- 
isation. By J. C. Firth, Author of 
'Luck' and * Our Kin across theSeau' 
Crown 8vo. 6f. 

FITZWYGRAM, — Horses and 

Stables. By Major-General Sir F. 
FiTZWYORAM, Bart. With 19 pages of 
Illustrations. 8vo. 5^. 

FORBES.— K Course of Lectures 

on Electricity, delivered before the 
Society of Arts. By Gborge Forbes. 
With 17 Illustrations. Crown 8vo. 51. 

FORD.—Thit Theory and Prac- 
tice of Archery. BythelateHoRACi 
Ford. New Edition, tborooghly Revised 
and Re- written by W. Burr, KLA. With 
a Preface by C. J. Longman, ^UA. 
F.S.A. 8vo. 14^. 

/OX— The Early History of 
Charles James Fox. By the 

Right Hon. Sir G. O. Trktelyan, Bait. 
Library Edition, 8to. iSf. 
Cabinet Edition, cr. 8vo. 6f. 

FRANCIS.— K Book on Angling; 

or. Treatise on the Art of Fishing in every 
branch; including full Illustrated List 
of Salmon Flies. By Francis Francis. 
Post 8vo. Portrait and Plates, ly. 

FREEMAN. — The Historical 
Geography of Europe. By e. a. 

Freeman. WithdsBAaps. 2 vols. 8vu 
31/. 6£U 

FROUDE (Janus ^.)— Works by. 
The History of England, from 

the Fall ol Wolsey to the Defeat of the 
Spanish Armada. 12 vols, crown %^. 

Short Studies on Great Sub- 
jects. 4 vols, crown 8vo. 24s, 

Cesar : a Sketch. Crown Svo. 3^. 6rf. 

The English in Ireland in the 
Eighteenth Century. 3 vols. 

crown 8va l8x. 

Oceana ; or, England and Her 

Colonies. With9 illustrations. Crown 
8vo. 2s, boards, 2f. 6d. doth. 

The English in the West Indies; 

or, the Bow of Ulysses. With 9 
Illustrations. Crown 8vo. 2s. boards, 
2s, 6d. cloth. 

The Two Chiefs of Dunboy; 

an Irish Romance of the Last Century. 
Crown 8vo. 6s, ' 

Thomas Carlvlei a History of his 

Life, 1795 to *o35. 2 vols, crown Svo. 
7 J. 1834 to 1 881. 2 vols, crown 8vo. 7j^ 


the Diseases Classified bjr the 
Registrar-General as Tabes 
Mesenterica. By w. t. gairdner, 
M.D. LL.D. On the Pathology 
of Phthisis Pulmonalis. By 

Joseph Coats, M.D. With 28 lUustra- 
ations. 8vo. I2J. 6</. 

GALLOWAY.— 'Doit Fundamen- 
tal Principles of Chemistry 

Practically Taught by a New Method. By 
Robert Galloway. Cr. 8vo. 6f. td, 

GANOT^ {Professor) — yUov^iS by. 
Translated by £. Atkinson, Ph. D. F.C.S. 

Elementary Treatise on Phy- 
sics. With 5 Coloured Plates and 923 
Woodcuts. Crown Svo. 15J. 

Natural Philosophy for General 

Readers. With 2 Plates, 518 Wood- 
cuts, and an Appendix of Questions. 
Crown 8vo. yj. ^. 

GARDINER {Samuel Rawson)— 
Works by. 

History of England, from the 

Accession of James I. to the Outbreak 
of the Civil War, 1603- 1642. 10 vols, 
crown Svo. price ts, each. 

A History of the Great Civil 
'War, 1642-1649, (3 vols.) VoL I. 

1642-1644. With 24 Maps. 8va 2ij. 
Vol.11. 1644-1647. With21Maps.8vo.24j. 

GARROD {Sir Alfred Baring)— 
Works by. 

A Treatise on Gout and Rheu- 
matic GpUt (Rheumatoid Arthritis). 
With 6 Plates, comprising 21 Figures 
(14 Coloured), and 27 Illustrations en- 
graved on Wood. Svo. 2ij. 

The Essentials of Materia 
MedicaandTherapeutics. New 

Edition, revised and adapted to the New 
Edition of the British Pharmacopoeia, by 
Nestor Ti&a&d, M. D. Cr. Svo. I2x. 61/. 

GIBSON.— h Text-Book of Ele- 
mentary Biology. ByR.T.HARVEY 

Gibson, M.A. With 192 Illustrations. 
Fcp. Svo. 6/. 

GOETHE.— FausL A New Transla- 
tion chiefly in Blank Verse ; with Intro- 
duction and Notes. By James Adby 
Birds. Crown Svo. 6s. 

Faust The Second Part A New 
Translation in Verse. By James Adey 
Birds. Crown Svo. 6s. 

GRA K -— Anatomy, Descriptive 
and Surgical. By henry gray, 

F. R. S. With 569 Woodcut Illustrations, 
a large number of which are coloured. 
RoyadSvo. 36^. 

GREEN-ThtWoTks of Thomas 

Hill Green. Edited by R. L. Nettle- 
ship (3 vols.) Vols. I. and II.— 
Philosophical Works. Svo. i6s. each. 
Vol. III.--Miscellanies. With Index to 
the three Volumes and Memoir. Svo. 21s. 

The Witness of God and Faith : 

Two Lay Sermons. By T. H. Green. 
Fcp. Svo. 2S. 

GREFILLE.^A Journal of the 
Reigns of King George IV. 
King William I v. and Queen 
Victoria. By c. c. f. Grevillb. 

Edited by H. Reeve. Svols. Cr.8vo.6f.ea. 

G WILT— An Encyclopaedia of 

Architecture. By Joseph Gwilt, 
F.S.A. Illustrated with more than 1,700 
Engravings on Wood. Svo. 52J. 6d. 

HAGGARD, — lA{^ and its 

Author : an Essay in Verse. By Ella 
Haggard. With a Memoir by H. Rider 
Haggard, and Portrait. Fcp. Svo. ^.6d. 

HAGGARD {ff. Rider)— Works 

She. With 32 Illustrations by M 
Greiffenhagen and C. H. M. Kere. 
Crown Svo. 3/. 6tl, 

Allan Quatermain. With 31 Il- 
lustrations by C. H. M. Kerr. Crown 
Svo. 3J. bd, 

Maiwa's Revenge; or, the War 

of the Little Hand. Crown Svo 2x. 
boards ; 2s. 6d. cloth. 

Colonel Quaritch, V.C. A Novel. 

Crown Svo. y. 6d. 

Cleopatra : being an Account of the 
Fall and Vengeance of Harmachis, the 
Royal Egyptian. With 29 FuU-page 
lUuslrations by M. Greiffenhagen and 
R. Caton WoodviUe. Crown Svo. 6s. 

Beatrice. A Novel Cr. 8vo. 6s. 



Catalogue or Gbnbmal and SasNTHTC Boon 

JUKES (Andreuf)—yf/OTks by. 
The New Man and the Eternal 

Life, Crown 8vo. 6/. 

The T3rpes of Genesis. Crown 

8vo. 7/. 6d. 

The Second Death and the Res- 
titution of AU Things. Crown 

The Mystery of the Kingdom. 

Crown 8vo. 3^. 6d, 

The Names of God in Holy 

Scripture; a Revelation of Hii Nature 
and Relationships. Crown 8vo. 41. 6d, 

JUSTINIAN.-^T\Mt Institutes of 

iUStinian ; Latin Text, chiefly that of 
[uschke, with English Introduction. 
Translation, Notes, and Snmmaiy. By 
Thomas C. Sanda&s, M.A. 8vo. i8r. 

KALISCH{M. Jl/.)— Works by. 

Bible Studies. Part I. The Pro- 
phecies of Balaam. 8vo. lOf. 61/. Part 
II. The Book of Jonah. 8vo ios.6d. 

Commentary on the Old Testa- 
ment ; with a New Translation. Vol. I. 
Genesis, 8vo. i8j. or adapted for the 
General Reader. I2j. Vol. II. Exodus, 
15J. or adapted for the General Reader, 
I2J. Vol. III. Leviticus, Part I. 151. or 
adapted for the General Reader, 8x. 
Vol. IV. Leviticus, Part IL 151. or 
adapted for the General Reader, 8r. 

Hebrew Grammar. With Exer- 

cises. Part I. 8vo. I2J. W. Key, $/. 
Part IL I2X. 6d, 

XANT{Immanu€l)—V/OTks by. 

Critique of Practical Reason, 
and other Works on the 

Theory of Ethics. Translated 
by T. K. Abbott, B.D. With Memoir. 
8vo. I2s,6d, 

Introduction to Logic, and his 
Essay on the Mistaken Sub- 
tilty of the Four Figures, 

Translated hy T. K. Abbott. Notes by 
S. T. Coleridge. Svo. 6s, 

KENDALL (il/iiy)— Works by. 
From a Garrett. Crown Svo. 6j. 
Dreams to Sell ; Poems. Fcp. 

Sto. 6j. 

'Such is Life': a Novel. Crown 
8vo. ts. 

K/LUCK—Hwidbook to Mill's 

System of Logic. By the Rev. a. 
IL KlLUCK, M. A. Crown 8vo. Jf» 6d, 

KOLBE.^A Short Text-Book of 
Inorganic Chemistiy. By Di. 

Hermann Kolbk. With 66 Iliustnt- 
tions. Crown 8va 7x. 6d, 

LADD.—Bltmtnts of Physiolo- 
gical Psjchology: By Georgs 
T. Ladd. 8vo. 21J. 

LANG (Andrfw)^V/0Tks by. 
Custom and Myth: Studiesof Early 

Usage and Belief. With 15 Dlustxatioiis. 
Crown 8to. yj. 6d. 

Books and Bookmen. With 2 

Coloured Plates and 17 Illustrations. Cr. 
8vo. 6s. 6d, 

Grass of Parnassus. A Volume 

of Selected Verses. Fcp. 8vo. 6s, 

Letters on Literature. Crown 

8vo. dr. 6d. 

Old Friends : Essays in Epistolary 

Parody, dr. 6d, 

Ballads of Books. Edited by 

Andrew Lano. Fcp. 8vo. dx. 

The Blue Fairy Book. Edited by 

Andrew Lan& With numerous Il- 
lustrations by II. J. Ford and G. P. 
Jacomb Hood. Crown 8vo. df. 

The Red Fairy Book. Edited by 

Andrew Lang. With numerous Illos- 
trations by H. J. Ford and Lancelot 
Speed. Crown ovo. dr. 

LA FIGEKIE.^ Cardinal Lavi- 
gerie and the African Slave 
Trade, i vol 8yo. 14s. 

LECKV( W. E. ^)-Works by. 
History of England in the 
Eighteenth Century. 8vo.Vois. 

I. & II. 1700-1 76a 3dx. Vols. III. 
&IV. 1760-1784. 36X. V0IS.V. &VI. 
1784-1793- 36J. 

The History of European Morals 
from Augustus (o Charle- 
magne* 2 vols, crown 8vo. idr. 

History of the Rise and Influ- 
ence of the Spirit of Ra- 
tionalism in Europe. % vols. 

ciowB 8va i6s. 



B.C. 1887, A Ramble in 
British Columbia. By j. a. 

Lees and W. J^ Clutterbuck. With 
Map and 75 Illustrations, Crown Svo. dr. 

ZE WES.— The History of Philo- 
sophy, from Thales to Comte. By 
Gbo&oe Henry Lewes. 2vo1s« Svo. 32^. 

Light through the Crannies.— 

Parables and Teachings from the other 
Side. First Series. Cr. 8vo. i^. swd. ; 
1^. 6d. doth. 

LIVEING (Robert)-yNov\i% by. 

Handbook on Diseases of the 
Skin. Fcp Svo. 5j. 

Notes on the Treatment of Skin 
Diseases. iSmo.'3J. 

LLO YD,—h Treatise on Magnet- 
ism, General and Terrestrial. By H. 
LLOYDy D.D. D.C.L. 8va lOf. td. 

ZLO YD.— The Science of Agri- 
culture. By F.J. Lloyd. Svo. I2j. 

LONGMAN\Fredenck »^HWorks 

Chess Openings. Fcp. Svo. 2s, 6d. 

Frederick the Great and the 
Seven Years' War. Fcp. Svo. 

2J. 6d, 

A New Pocket Dictionary of 
the German and English 

Languages. Square iSmo. 2i. 6^. 

Longman's Magazine. Published 

Monthly. Price Sixpence. 
Vols. 1-15, Svo. price 51. each. 

Longmans' New Atlas. Political 

and Physical. For the Use of Schools 
and Private Persons. Consisting of 40 
Quarto and 16 Octavo Maps and Dia- 
grams, and 16 Plates of Views. Edited 
by Geo. G. Chisholm, M.A. B.Sc. 
Imp. 4to. or imp. Svo. I2r. 6d, 

LONGMORE {Surgeon-Geneml Sir 

2:)— Works by. 
Gunshot Injuries; their History, 

Characteristic Features, &c With 5S 
Illustrations. Svo. 31^. 6d, 

The Illustrated Optical Manual; 

or, Handbook of instructions for the 
Guidance of Surgeons in Testing Quality 
and Range of Vision. With 74 Illus- 
trations. Svo. 14J. 

LOUDON (/ C.)— Works by. 
Encvclopxdia of Gardening. 

With 1,000 Woodcuts. Svo. 21s. 

Encyclopaedia of Agriculture: 

the Laying-out, Improvementi and 
Management of Landed Property. With 
1 , 100 Woodcuts. Svo. 2 ix. 

Encyclopaedia of Plants; the 

Specific Character, &c. of all Plants found 
in Great Briuin. With 12,000 Wood- 
cuts. Svo. 42J. 

LUBBOCK— The Origin of Civil- 
isation and the Primitive Condition 
of Man. By Sir J. Lubbock, Bart. 
M.P. With 5 Plates and 20 lUustiatioas 
in the text. Svo. I&r. 

L K^ZZ.— The Autobiography[of a 

Slander. By Eona Lyall, Author 
of * Donovan,' &c. Fcp. Svo. is, sewed. 

L]rra Germanica ; Hymns Trans- 
lated from the German by Miss C. 

WlNKWORTH. Fcp. Svo. SJ. 

MAC AULA Y.-VJov)^% and Life 

History of England from the 
Accession of James the 

Popular Edition, 2 vols, crown Svo. 5/. 
Student's Edition, 2 vols, crown Svo. I2x. 
People's Edition, 4 vols, crown Svo. i6/. 
Cabmet Edition, S vols, post Svo. 4&r. 
library Edition^ 5 vols. Svo. £4. 

Critical and Historical Essays, 
with Lays of Ancient Rome^ 

in I volume : 
Popular Edition, crown Svo. 2s, 6d, 
Authorised Edition, crown Svo. 2s» 6d, or 

ys, 6d, gilt edges. 

Critical and Historical Essays]: 

Student's Edition, i vol. crown Svo. 6s. 
People's Edition, 2 vols, crown Svo. 8x. 
Trevelyan Edition, 2 vols, crown Svo. gs, \ 
Cabinet Edition, 4 vols. Dost Svo. 24/. 
library Edition, 3 vols. Svo. 36r. 

ICofUsmutf 0n tuxt page. 


Cataloguu op Gbhbmal Asn SctENTmc Books 

MAC AULA Y~-VJ0TkB and Life 
of LOJ^D MACA VLA Y^omd. 

Essays which may be had sqMur- 
ately price 6dl each lewed, i/. each doth : 

Addison and Walpole. 

Frederick the Great. 

Croker's BosweU's Tohnson. 

HalUm's Constitutional Htstoiy. 

Warren Hastingi. (>/. sewed, &/. doth.) 

The Earl of Chatham (Two Essays), 

Ranke and Gladstone. 

Milton and MachiaTeUi« 

Lofd Bacon* 

Lord Clive. 

Lord Byron, and The Comic Dramatists of 
the Restoration. 

The Essay on Warren Hastings annotated 

by S. Halbs, ij. dd. 
The Essay on Lord Give annotated by 


Speeches : 

People's Edition* crown 8to. 31. &/• 

Miscellaneous Writings : 

People's Edition, i vol. crown 8vo. 4r. 6</. 
library Edition, a yolt. 8vo. au. 

Lays of Ancient Rome, &c. 

Illustrated by G. Sdiarf; fcp. 4to. lor. dd. 
— — ^— — ^— Bijoa Edition, 

i8mo. 2x. td, gilt top. 
Popnlar Edition, 

fq). 4to. dd, sewed, i/. doth. 
Illustrated by J. R. Weguelin, crown 8Ta 

y. 6d, doth extra, gilt edges. 
Cabinet Edition, post Ovo. 3;. 6</. 
Annotated Edit. fcp. 8vo. li. sewed, u,6d, d. 

Miscellaneous Writings and 
Speeches : 

Popular edition, x vol. crown 8vo. 2s, &/. 

Student's Edition, in i vol. crown 8vo. 6x. 

Cabinet Edition, induding Indian Penal 
Code, Lays of Andent Rome, and Mis- 
cellaneous Poems, 4 vols, post 8vo. 24X. 

Selections from the Writings 
of Lord Macaulay. Edited, with 

Occasional Notes, by the Right Hon. Sir 
G. O. TuvBLYAN, Bart. Crown 8vo. 6x. 

Complete Works of Lord Ma- 

Library Edition, 8 vols. 8vo. £$. $s. , 
Cabinet Edition, 16 vols, post 8vo. £^ i6x. 

The Life and Letters of Lord 

Macaulay. By the Right Hon. Sir 

G. O. Trbvelyan, Bart. 
Popular Edition, i vol. crown 8vo. aj. 6d* 
Student's Edition, I voL crown 8vo. 6x. 
Cabinet Edition, a vols, post 8vo. I2x. 
Library Edition, 2 vols. 8vo. 36/. 

MACDONALD ((?«?.) — Works 

"^ UnspcAen Sermons. First and 

Second Series. Crown 8vo. 3/. 6dl each. 
Third Series. Crown 8vo. ^s, 6d. 

The Miracles of Our Lord. 

Crown Svo.. ^r. 6d, 

A Book of strife, in the Form 
of the Diary of an Old Soul : 

Poems. lamow fie 

Lectures on Harmony. 8to. 12s. 
Addresses and Lectures. Cr. 

8vo. 6f . 6d. 

MACLEOD {Henry Z>.) — Works 

The Elements of Banking. 

Crown Svo. 5/. 

The Theory and Practice of 

Banking. Voi.L8vo.1ar.VoLn.14j. 

The Theory of Credit 8vo. 

VoL L V' 6d.; VoL IL Part L 4^. 6^1 

Mcculloch.— Tht Dictionaiy 

of Commerce and Commerdal Navi- 
gation of the late J. K. McCulloch. 
ovo. with 1 1 Maps and to Charts, price 
63r. doth, or yoir. half-boand in rassia. 

McDOUGALL. — Memoirs of 
Francis Thomas McDougall, 

D.C.L. F.R.C.S. sometime Bishop 
of Labuan and Sarawak, and of Harriette 
his Wife. By her Brother, Charles J. 
BUNYON. Svo. 14J. 

Mademoiselle Mori: a Tale of 

Modem Rome. By the Author of * The 
Atelier da Lys.' Crown 8vo. 2J. 6^. 

MALMESBURY. — VL^mQ]x% of 
an Ex-Minister. By the Eari of 

Malmesbd&y. Crown 8vo. 7/. 6^ 

PHILOSOPHY {Stmyhurst 

Logic. By Richard F. Clarkle, S.J. 

Crown Svo. 5J-. 

First Principles of Knowledge. 

By John Rickaby, S.J. Crown 8vo, 5J. 

Moral Philosophy (Ethics and 
Natural Law). By Joseph Rick- 
ABY, S.J. Crown Svo. 5/. 

{CoHtinuid on next pagt. 



Manuals of Catholic Philosoi^y 

(Stonyhurst Series) — continued* 

General Metaphysics. By John 

RlCKABY, SJ. Crown 8va 5j. 
' Ps^hology. By Michael Maher, 
SJ. Crown 8vo. 6/. 6df. 

Natural Theology. By Bernard 

fioEDDER, S.J. Crown 8vo. 6r. td, 
[Nearly ready. 

A Manual of Political Economy. 

By C. S. Devas, £s^. M.A. Examiner 
in Political Economy m the Royal Uni» 
versity of Ireland. 6i,6d. [In preparation, 

JlM^r/iV:— Navigation and Nau- 
tical Astronomy. Compiled by 
StaflF-Commander W. R. Martin, R.N. 
Royal 8vo. i&f. 

MARTINEAU O^isx) — Works 

Hours of Thought on Sacred 

Things. Two Volumes of Sermons. 
3 vols, crown 8vo. *]s, 6^. eacb. 

Endeavours after the Christian 

Life. Discourses. Crown 8vo. 7j. 6</. 

The Seat oif Authority in Re- 
ligion. 8vo. I4f. 

MATTHEWS {Brander)—yNovks 
Pen and Ink. Papers on Subjects 

of More or Less Importance. Cr. 8vo. 5^. 
A Family Tree ; and other Stories. 

Crown 8vo. dr. 

Biographical Treasury. With 

Supplement brought down to 1889, by 
Rev. Jas. Wood. Fcp. 8vo. dr. 

Treasury of Natural History; 

or, Popular Dictionary of 2^1ogy. Fcp. 
8vo. with 900 Woodcuts, dr. 

Treasury of Geography, Physical, 

Historical, .Descriptive, and Political. 
With 7 Maps and 16 Plates. Fcp. 8 vo. 6x. 

Scientific and Literary Trea- 
sury. Fcp. 8vo. 6s, . 
Historical Treasury: Outlines of 

Universal History, Separate Histories of 
all Nations. Fcp. 8vo. dr. 

Treasury of Knowledge and 
Libraxyof Reference. Comprising 

an English Dictionary and Grammar, 
Univeried Gazetteer, Classical Dictionary, 
Chronology, Law Dictionary, &c. Fcp. 
8vo. 6s, 



The Treasury of Bible Know- 
ledge. By the Rev. J. Atrb, M.A. 
With 5 Maps, 15 Plates, and 300 Wood- 
cuts. Fcp. 8vo. dr. 

The Treasury of Botany. 

Edited by J. Lindley, F.R.S. and 
T. Moore, F. L. S. With 274 Woodcuts 
and 20 Steel Plates. 2 vols. fcp. 8vo. ^I2j. 

MAXMULLER (i?!)— Works by. 

Selected Essays on Language, 
Mythology and Religion. 2 vols. 

crown 8vo. \6s. 

Lectures on the Science of Lan- 
guage. 2 vols, crown 8vo. idr. 

Three Lectures on the Science 

of Language and its place in General 
Education. Crown 8vo. 2J. 

India, What can it Teach Us? 

A Course of Lectures delivered before the 
University of Cambridge. 8vo. I2J. 6d, 

Hibbert Lectures on the Origin 
and Growth of Religion, as 

illustrated by the Religions of India. 
Crown 8vo. fs, 6d. 

Introduction to the Science of 

Religion ; Four Lectures delivered at 
the Roya^ Institution. Crown 8vo. 7^. 6d, 

Natural Religion. The Gifford 

Lectures, delivered before the University 
of Glasgow in 1888. Crown 8vo. lor. 6d, 

The Science of Thought. 8va 2 ix. 

Three Introductory Lectures on 

the Science of Thought. Svo. 

2/. 6d, 

Biographies of Words, and the 
HomeoftheAr]ras. Cr8vo.7J.d/. 

A Sanskrit Grammar for Be- 

§inner& New and Abridged Edition. 
y A. A. MacDonell. Crown 8vo: dr. 

MAY,—Tti!t Constitutional His- 
tory of England since the Accession 
of George III. 1760-1870. By the Right 
Hon. Sir Thomas Erskine May, K.C.B. 
3 vols, crown Svo. I&r. 

MEATH{TheEarlof)—V7ox)si^ by. 

Social Arrows : Reprinted Articles 
on various Social Subjects. Cr. 8vo. 5/. 

Prosperity or Pauperism ? Phy- 
sical, Industrial, and Technical Training. 
(Edited by the Earl of Meath). 8vo. 5^. 

Catalogue of Gessmal asd ScmntFic Books 

MELBOURNE.'-Th^ Melbourne 

Papers : being a Selection from Docu- 
menu in the possession of Earl Cowpcr, 
K.G. Edited by Lloyd C. Sanders, 
B.A. 8vo. i8j. 
MELVILLE {G.J. lVhyte)^Yloyf€is 
by. Crown ^vo. is. each, boards; 
i^. 6d. each, doth. 

The GUdiators. Holmby House. 

The Interpreter. Kate Coventiy. 

Good for Nothing. Digby Grand. 

The Queen's Maries, General Bounce. 

MENDELSSOHN.— Th^ Letters 
of Felix Mendelssohn. Trans- 

lated by Lady Wallace. 2 vols, crown 
8vo. lox. 
MERIVALE {The Very Rev. Chas.) 
Works by. 
History of the Romans under 

the Empire, cabinet Edition, 8 vols, 
crown 8vo. 481. 
Popular Edition, 8 vols, crown 8va zs. oa. 

The Fall of the Roman Republic : 

a Short History of the Last Century of 
the Commonwealth. i2mo. yj. od. 

General History of Rome from 

B.C. 753 to A.D. 476. Cr. 8vo. 71. 6d. 

The Roman Triumvirates. With 

Maps. Fcp. 8vo. ax. 6d. 

MEYER.— Modern Theories of 

Chemistry. By Professor Lothar 
Meyer. Translated by P. Phillips 
Bedson, F.CS. ; and W. Carleton 
Williams, F.C.S. 8vo. i8x. 

;j«z:z.-Analysis of the Pbeno- 
mena of the Human Mmd. 

By James Mill, a vols. 8vo. aSs. 
MILL{/oAn Stuart)— Works by. 
Principles of Political Economy. 

Library Edition, 2 vols. 8vo. sor. 
People's EdiUon, I vol. crown 8vo. Jr. 

A System of Logic. Cr. 8vo. 51. 
On Liberty. Crown 8va is. 4^ 
On Representative Government 

Crown 8vo. 2s. 

Utilitarianism. 8va 5^. 
Examination of Sir William 
Hamilton's Philosophy. 8vo.i6x. 

Nature, the Utility of Religion, 

and Theism. Three Essays. 8vo. 5/. 

MILLER (IV. AUeH)—V/orks by. 
The Elements of Chemistry, 

Theoretical and PiacticaL With Addi- 
tions, by H.Maclbod.F.CS. 3vol».8vo. 
Vol. I. Chemical Physics, i6t. 
Vol. II. Inorganic Chemistry, 24s, 
VoL III. Organic Chemistry. 31J. 6d. 

An Introduction to the Study 
of Inorganic Chemistry. With 71 

Woodcuts. Fcp. 8vo. 31. 6^. 

MITCHELL.— A Manual of Prac- 
tical Assaying. ByJOHNMlTCHEXX. 
Revised by W. Crookbs. 8vo. Wood- 
cats, 3U. 6ii. 

MirCHELJ:. — Dissolution and 
Evolution and the Science of 

Medicine. ByCPiTFiELD Mitchell. 
8vo. i6tf. 
MOLElSlVORTIf.—yLBxrying and 
Giving in Marriage: a Novel. 

By Mrs. MOLESWORTH. Fcp. 8vo. 2x. 6/. 

MOZLE v.— TheV/ord. BytheRev- 

T. Mozley, Author of * Reminiscences 
of Oriel College and the Oxford Move- 
menL' Crown 8vo. 7^. 6d. 

MVLHALL.'-Historj of Prices 
since the Year 185a By Michael 
G. MULHALL, Crown 8vo. 6s. 

MURCHISON{aas.)-V/orks by. 

A Treatise on the Continued 

Fevers of Great Britain. With 

numerous Illustrations. 8vo. 251. 

Clinical Lectures on Diseases 
of the Liver, Jaundice, and 
Abdominal Dropsy. With 43 

Illustrations. 8vo. 24J. 

MURDOCK.— The Reconstruc- 
tion of Europe : a Sketch of the 
Diplomatic and Military History of 
ContinenUl Europe, from the Rise to the 
Fall of the Second French Empire. By 
Henry Murdock, Crown 8vo. 91. 

MURRAY.— h Dangerous Cats- 
paw : a Story. By David Christie 
Murray and Henry Murray. Cr. 8vo. 
zs, 6d. 

MURRA K— Gobi or Shamo. By 

G. G. A. Murray. Crown 8vo. 2j. 
boards; 2J. 6^. cloth. 

Darrie:aStory. ByCHRisxiEMuRRAY 
and Hrnry Herman, Crown 8vo. 2x. 
boards ; 2j. (>d, doth. 

PUBUSHBD Br Messrs. Lokcmans^ Gxbbn, 6* Co. 


NAPIER, ^Tht Life of Sir 

{oseph Napier, Bart. Ex- 
•ord Chancellor of Ireland. 

By Alex. Charles Ewald, F.S.A. 
With Portrait. 8vo. 15/. 

NAFI£R.— The Lectures,Essays, 
and Letters of the Right Hon. 
Sir Joseph Napier, Bart latc 

Lord Chancellor of Ireland. AVith an 
Introduction by his Daughter. Forming 
a Supplement to * The Life.* With Por- 
trait. 8vo. I2s, 6d. 

J)ri:SBIT{E.)—'Works by. 
Lays and Leg:ends. Cr. 8vo. 5^. 
Leaves of Life ; Verses. Cr. 8vo. 55. 
NEWMAN.^On the Diseases of 
the Kidney Amenable to Sur- 
gical Treatment. By David 
Mewman, M.D. 8vo. i6r. 

NE WMAN (Cardinal)— V/orkS by. 
Apologia pro Vitk Sua. Cabinet 

Edition, cr. 8vo. 6j. Cheap Edition, y.6d. 

The Idea of a University defined 

and illustrated. Crown 8vo. 7s, 
Historical Sketches. 3 vols. 

crown 8vo. 6s, each. 

The Arians of the Fourth Cen- 
tury. Crown 8vo. 6t, Cheap Edition, 
crown 8vo. 3J. 6d. 

Select Treatises of St Athan- 

asiuS 'in Controversy with the Arians. 
Freely Translated. 2 vols. cr. 8vo. 15^. 

Discussions and Arguments on 
Various Subjects. Crown 8vo. 6j. 

An Essav on the Development 
of Christian Doctrine. Cabinet 

Edition, crown 8vo, 6s, Cheap Edition, 
crown 8vo. 3/. 6^. 

Certain Difficulties felt by 
Anglicans in Catholic Teach- 
ing Considered. Vol. i, crown 

8vo. 7j. 6</. ; VoL 3, crown 8vo. 5/. 6d, 

The Via Media of the Anglican 

Church, illustrated in. Lectures, &c. 
2 vols, crown 8vo. 6s, each. 

Essays, Critical and Historical. 

Cabinet Edition, 2 vols, crown 8vo. I2s. 
Cheap Edition, 2 vols, crown 8vo. Js, 

Essays on Biblical and on Eccle- 
siastical Miracles. Crown 8vo.6j. 
An Essay in Aid of a Grammar 

of Assent. Crown 8vo. p. 6d. 

NEWMAN (Cardi7ial)-AN orks by 


. Present Position of Catholics in 

England. Crown 8vo. Js, 6d. 
Callista : a Tale of the Third Gen- 
tury. Cabinet Edition, crown 8vo. 6j. 
Cheap Edition, crown 8vo. y, 6d. 

The Dream of Gerontius. i6mo. 

6d, sewed, is, cloth. 

Verses on Various Occasions. 

Crown 8vo. 6j. Cheap Edition, crown 
8vo. 3J. 6d. 

NICirOZSON—ToxdLr iSiRomance. 
By J. Shield ]^icholson, Author of 
*Thoth,' &c. Crown bvo. 6s. 

NOBLE.-^Honrs with a Three- 
inch Telescope. By Captain W. 
Noble. With a Map of the Moon. Cs. 
8vo. 4s. 6d, 

NORRIS.'-yiTS. Fenton : a Sketch. 

By W. E. NoRRis. Crown 8vo. dr. 

NORTHCOTT. — 'Lzthts and 

Turning, Simple, Mechanical, and 
Ornamental. By W. H. Northcotf. 
With 338 Illustrations. 8vo. iZs. 

O'BRIEN— V/hen we were Boys : 

a Novel. By William O'Brien, M.P. 
Crown 8vo. 6s, 


Madam. Cr. 8vo. is. bds. ; is. 6d. cI. 

In Trust— Cr. 8vo. is. bds. ; 1^. 6d. cl. 

Lady Car : the Sequel of a Life. 

Crown 8vo. 2x. 6d, 

OLIVER.— Astronomy for Ama- 
teurs : a Practical Manual of Telescopic 
Research adapted to Moderate Instru- 
ments. Edited by J. A, Westwood 
Oliver. With several Illustrations; 
Crown 8vo. 7j. 6d. 

OWEN—h Manual of Anatomy 
for Senior Students. By Edmund 

Owen, M.B. F.R.C.S. With 210 
Illustrations. Crown 8vo. I2s, 6d, 

PAGET (Sir fames)— V/orks by. 

Clinical Lectures and Essays. 

Edited by F. Howard Marsh. 8vo. 15X. 

Lectures on Surgical Patho- 
logy. 8vo. with 131 Woodcuts, 21s, 

PASTEUR.— l^ovLis Pasteur, his 

Life and Labours. By his Son-in-Law. 
Translated from the French by Lady 
Claud Hamilton. Crown 8vo. 71. 6d. 


Catalogub of Gmsmmal and SciBsnnc Booa 

A^KifM— Industrial Chemistry. 

a Manual for Manafactoren and for Col- 
leges or Technical Schools; aTraoslation 
of Payen*s *Pr^is de Chimie Indus- 
trielle' Edited by a H Paul. With 
698 Woodcuts* Medium 8vo. 4^. . ^_ 

FA YN Oww)— Novdfc by. 
The Luck of the Darrells. Cr. 

8va II. boaids ; is, 6d, doth. 

Thicker than Water. Crown 8va 

u. boaids ; \s. 6d, doth. 

J>ENNELL.—Out Sentimental 
Journey through France and 

Italy. By Joseph and Elizabeth 
Robins Pennell. With a Map and 
120 lUustialions by Joseph PennelL 
Cr. 8vo. 6f. 

PERRING (SrP/Lt'Iipy-Works by. 
Hard Knots in Shakespeare. 

Svo. 7^. 6d. 

The 'Works and Days' of 

Moses. Crown 8vo. 3/. &/. 


a Legend of the Lone Mountain. By C. 
PuiLLiPPS-WoLLEY, Author of * Sport in 
the Crimea and Caucasus '&c. With 13 
Illustrations by H. G. Willink. Crown 
8vo. 6r. 

POLE.— The Theory of the Mo- 
dem Scientific Game of 

Whist. By W. Pole, F.R.S. Fcp. 
8vo. 2f. 6d, 

PORTER.— The History of the 
Corps of Royal Engineers. 

By Major-General Whitworth Porter, 
R.E. a vols. 8vo. 361. 

the Restoration to the Revo- 
lution, i66o-i69a By John P. 
Prbndergast. 8yo. 5s. 

PROCTOR (R. ^.)— Works by. 
Old and New Astronomy. 12 

Parts, 2s, 6d, each. Supplementary Sec- 
tion, IS, Complete in i vol. 4to. 36;. 
[In course of puhlkatum^ 

The Orbs Around Us ; a Series of 

Essays on the Moon and Planets, Meteors 
and Comets. With Chart and Diagrams. 
Crown Svo. 5j. 

Other Worlds than Ours; The 

Plurality of Worlds Studied tinder the 
Liffht of Recent Scientific Researches. 
With 14 Illustrations. Crown 8vo. 5/. 

PROCTOR {R. >f.)— Works by— 


The Moon ; her Motions, Aspects. 
Sceneiy, and Physical Condition* With 
Plates, Charts, Woodcuts, ac. Cr.8vo.5j. 

Universe of Stars; Presoitiiig 

Resea r ches into and New Views respect- 
ing the Constitntioa of the Heavens. 
With as Chaits and 2a Diagnms. Svo. 

Lai^ffer Star Atlas for the Libraxy, 

in 12 Circular Maps, with Introdnctioa 
and 2 Index Pages. Folio, I5r. or Maps 
only, I2r. 6d. 

The student's Atlas* In Twelre 

Circular Maps on a Uniform Projection 
and one Scale. 8va 5/. 

New star Atlas for the Library, 
the School, and the Observatory, in X2 
Circular Maps. Crown 8vOb 5/. 

LightSciencefor Leisure Hours; 

Familiar Essavs on Scientific Sid]jects. 
3 vols, crown 8vo. 5/. each. 

Chance and Luck ; a Discussion of 

the Laws of Luck, Coincidences, Wagers, 
Lotteries, and the Fallacies of Gambling 
&C. Crown 8vo. 2j. boards ; 2x. €d, doth. 

Studies of Venus-Transits. With 

7 Diagrams and 10 Plates. 8vo. 5j. 

How to Play Whist : with the 
Laws and Etiquette of Whist 

Crown 8vo. 3;. 6^. 

Home Whist: an Easy Guide to 

Correct Play. i6mo is. 

The Stars in their Seasons. 

An Easy Guide to a Knowledge of the 
Star Groups, in X2 Maps. Roy. Svo. $s. 

Star Primer. Showmg the Starry 

Sky Week by Week, in 24 HonrlyMaps. 
Crown 4to. 2r. 6d, 

The Seasons Pictured in 48 Sun- 
Views of the Earth, and24Zod]acal 
Maps, &c Demy 4to. 5/. 

Strength and Happiness. With 

9 Illustrations. Crown 8vo. y. 

Strength : How to get Strong and 
keep Strong, with Chapters on Rowing 
and Swimming, Fat, Age, and the 
Waist. With 9 Illustrations. Crown 
8vo. 2J. 

Rough Ways Made Smooth. 

Familiar Essays on Scientific Subjects. 
Crown 8va 5r, 

Our Place Among Infinities. A 

Series of Essays contrastinf our little 
Abode in Space and Time with the Infi* 
nities Around us. Crown 8va Ss, 

[ConHtmed on next page. 



PROCTOR {R. ^.)— Works by— 


The Expanse of Heaven. Essays 

on the Wonders of the Firmament. Crown 
8vo. 5J. 

The Great Pyramid, Observa- 
tory, Tomb, and Temple. 

With niiisCrations. Crown 8to. ^r. 

Pleasant Wa]rs in Science. Crown 
Myths and Marvels of Astro* 

npmy. Crown 8vo. 5/. 

Nature Studies. By Grant Allen, 
A. Wilson, T. Foster, E. Clodd, and 
R. A. Pkoctor. Crown 8vo. y. 

Leisure Readings. By £. Clodd, 

A Wilson, T. Foster, A. C. Ranyard, 
and R. A Proctor. Crown 8vo. 51. 

PROTHERO.—TYat Pioneers and 
Progress of English Farming. 

By Rowland £. Prothero. Cr. 8vo. sr. 

-P^FCfi.— The Ancient British 

Church: an Historical Essay. By 
John Pryck, M. A Crown 8vo. 6r. 

Quain's Elements of Anatomy. 

Re-edited by Allen Thomson, M.f>. 
E. A SCHAFBR, F.R.S. andG. Dancer 
Thane. With upwards of 1,000 Illus- 
trations. 2 vols. 8vo. i8j. each. 

QUAIN.—K Dictionary of Medi- 
cine. By Varions Writers. Edited by 
R. QUAIN, M.D. F.R.S. &c. With 138 
Woodcuts. Medium 8to. 31/. 6</. cloth ; 
to be had also fai a vols. 341. doth. 

RAWLINSON.^Tht History of 

Phoenicia. By George Rawlin- 
SON, M.A Canon of Canterbury, &c. 
With numerous Illustrations. 8vo. 24J. 

READER (Emily -E.)— Works by. 
Echoes of Thought : a Medley of 

Verse. Fcp. 8vo. 5J. doth, gilt top. 

The Ghost of Brankinshaw and 

other Tales. With 9 Illustrations. Fcp. 
8to. 2j. 6d, doth extra, gilt edges. 

Voices from Flower-Land, in 

Original Couplets. A Birthday-Book and 
liUngnage of Flowers. i6mo. ls.6d, limp 
doth ; 2/. 6d, roan, gilt edges, or in vege- 
table vellum, gilt top. 

Fairy Prince FoUow-my-Lead ; 

or, the Magic Bracdet. Crown 8vo. 
2/. 6</. gilt edges; or ^r. 6</. vegetable 
vellum; gilt edges. 

Reply ( 

REEVE.— Co6kJtTj and House- 
keeping. By Mrs. Henry Reeve. 
With 8 Plates and 37 WoodcuU. Cr. 
8vo. 5j. 

Inns of Old Southwark, and 

their Associations. By William Rendle, 
F.R.CS. and Philip Norman, F.S.A 
With numerous niustrations. Roy.8vo.38x. 

(A) to Dr. Lightfoot's 

lyS. By the Author of ' Super* 
natural Religion.' I vol. 8vo. 6f. 

R/BOr.— The Psychology of At- 
tention. By Th. Ribot. Crown 
8vo, y, 

R/CIf.—A Dictionary of Roman 
and Greek Antiquities, with 

3,000 Woodcuts. By A Rick. Cr. 8vo. 
7/. 6^. 


Works by. 
The Commonhealtfa : a Series of 

Essays on Health and Felidty for Every* 
Day Readers. Crown 8vo. 6s. 

The Son of a Star : a Romance of 

the Second Century. Crown 8vo. 6s. 

National Health. Abridged from 

"The Health of Nations." A Review 
of the Works of Sir Edwin Chadwick, 
K.C.B. Crown 41. 6d. 

RILEY.-^MtkOs; or, the Mountain 
of the Monks. By Athklstan Riley, 
M.A F.R.G.S. With Map and 39 
Illustrations. 8vo. Zls. 

RIVERS {Thomas)— ^tJorliB by. 
The Orchard-House. With 25 

Woodcuts. Crown 8vo. $s. 

The Miniature Fruit Garden; 

or, the Culture of Pjrramidal and Bush 
Fruit Trees. With 33 Illustrations. Fcp. 
8vo. 4f. 

ROBERTS.— Gretk the Lan- 
guage of Christ and His 
Apostles. By Alsxandkk Roberts, 
D.D. 8vo. i&r. 

ie(?Gi?2!— Thesaurus of English 
Words and Phrases. ciassiSed 

and Arranged so as to. facilitate the Ex- 
pression of Ideas. By PsTBR M. ROGBT. 
Crown 8vo. lOr. 6d, 

Catalogue of Grmoul and Sasurmc Books 

RONALDS.— Tht Fly - Fishei^s 
Entomology. By Alfked Ronalds. 
With 20 Coloured Plates. 8vo, 14/. 

IiUSSELL.—h Life of Lord John 
RusscU (Earl Russell, K.G.). 

By Spencer Walpole. With 2 Por- 
traits. 2 vols. 8vo. 36i. 

SA RGANT. — Urban Rating : 

being an Inquiry irrto the Incidence of 
Local Taxation in Towns, with Spedai 
Reference to Current Proposals for 
Change. By Charles Henry Sar- 
CANT. »vo. 6j. 

SCHAFER.—Tht Essentials of 

Histology, Descriptive and Practical. 
By E. A. bCHAFER, F.R.S. With 281 
Illustrations. 8vo. 6f. or Interleaved 
with Drawing PRper, 8j. 6d. 

SC ff ELLEN. S^ctnxm Anal- 
ysis in its Application to Terrestrial 
Substances, and the Physical Constitu- 
tion of the Heavenly Bodies. By Dr. 
H. SCHELLBN. With 14 Plates and 
291 Woodcuts. 8vo. 3IJ. 6^. 

^COrr.— Weather Charts and 
Storm Warnings. By r. h.Scott, 

F.R.S. With Illustrations. Cr.8vo. 6j. 
SEEBOHM {Frederick) — VJovkz 

The Oxford Reformers— John 
Colet, Erasmus, and Thomas 

More ; a History of their Fellow-Work. 
8vo. 14J. 

The Era of the Protestant 

Revolution. With Map. Fcp.8vo. 
2/. fid. 

The English Village Commu- 
nity Examined in its Relations to the 
Manorial and Tribal Systems, &c 13 Maps 
and Plates. 8vo. its, 

SENNETT.—Tht Marine Steam 

Engine ; a Treatise for the use of 
Engineering Students and Officers of the 
Royal Navy. By Richard Sbnnbtt, 
Engineer-in-Chief of the Royal Navy. 
With 244 Illustrations. 8vo. 21/. 

SEWELL.—Stonts and Tales. 

By Eli^beth M. Sewell. Crown 8vo. 
IS. 6d. each, cloth plain j 2s. 6d, each, 
cloth extra, gilt edges :^ 

Amy Herbert. 
The Earrs Daughter. 
The Experience of Life. 
A Glimpse of the World. 
Cleve HaU. 
Katharine Ashton. 

Margaret Perdval. 
Laneton Parsonage. 

SHAKESPEARE. — Bowdler's 
Family Shakespeare, i vol. 

8vo. With 36 Woodcuts, I4i; or in 
6 voJs. icp. 8vo. 2ix. 

Outiines of the Life of Shake- 
speare. By J. O. Halliwell-Phil- 
Lipps. 2 vols. Royal 8vo. £u is, 

Shakespeare's True Life. By 

James Walter. With 500 Illustrations, 
Imp. 8vo. 2 1 J. 

SlfORT.^Sketch of the History 
of the Church of England 

to the Revolution of 16S8. By T. V. 
Short, D.D. Crown 8vo. 7/. 6d. 


She : a History of Adventure. By 
H. Rider Haggard. With 32 Illus- 
trations. 3/. 6d. 

Allan Quatermain. By H. Rider 

Haggard. With 20 Illustrations. Ss. 6if. 

Colonel Quaritch. V.C. : a Tale 

of CountryLife. By H. Rider Haggard. 
With Frontispiece and Vignette. 3s. 6ti 

Micah Clarke: His Statement. 
A Tale of Monmouth's Rebellion. By 
A. CoNAN Doyle. With Frontispiece 
and Vignette, jj. 6d. 

Petland Revisited. By the Rev. 

J. G. Wood. With 33 Illustrations. 
3s. 6J, 

Strange Dwellings: a Description 

of the Habitations of Animals. By the 
Rev. J. G. Wood. With 60 Illustrations. 
3^. &/. 
Out of Doors. Original Articles 
on Practical Natural History. By the 
Rev. J. G. Wood. With 11 lUustra 
tions. 3x. 6d. 

Familiar History of Birds. By 

the late Edward Stanley, D.D. Lord 
Bishop of Norwich. With 160 Woodcuts. 
3J. 6flf. 

Ei|:ht Years in Ceylon. By Sir 

S.W.Baker. With 6 Illustrations, y.^ti, 

Rifle and Hound in Ceylon. By 

Sir S. W. Bakkr. With 6 Illustrations. 

Memoirs of Major-General Sir 
Henry Havelock. By John 

Clark MARSHMAN. With Portrait. 3j.6c/. 

Visits to Remarkable Places, 

Old Halls, BattleEelds, Scenes Illustra- 
tive of Striking Passages in English 
History and Poetry. By William 
HowiTT. With 80 Illustrations. 31. 6^/. 
[Contintud on next pa^^ 





Field and Hedgerow. Last Essays 
of Richard Jefferies. With Portrait. 
3r. (>d. 

Story of Creation : a Plain account 
of Evolution. By Edward Clodd. 
With 77 Illustrations, y, 6d, 

Life of the Duke of Wellington. 

By the Rev. G. R. Glei^, M.A. With 
Portrait. 3^. 6d, 

History of the Romans under 

the Empire. By the Very Rev. 
Charles merivale, D.C.L. Dean of 
Ely. 8 vols. Each 3j. 6d, 

Cssar: a Sketch. By James A. 

Froude. 3j. 6d. 
Thomas Carlyle : a History of his 

Life. By J. A. Froude, M.A, Vols. 
I. and II. 1795-1835, 7J. Vols. III. 
and IV. 1834-1881, 7s. 

Apologia pro VitA Sua. By Car- 
dinal Newman. 3^. 6e/. 

Callista: a Tale of the Third Cen- 
tury. By Cardinal Newman. 3/. 6d, 

Essays, Critical and Historical. 

By Cardinal Newman. 2 vols, crown 

8V0. I2J. 

An Essay on the Development 
of Christian Doctrine. By Car- 

dinal Newman. Crown 8vo. ts, 
Electrical Engineering for 
Electric Light Artisans and 

Students. Jt*y W. Slingo and A. 
Brooker. With 307 Illustrations. Cr. 
8vo. xof. &/. 

SMITH {Gregory)^-VTZ. Angelico, 
and other Short Poems. By Gregory 
Smith. Crown 8va 4J. (>d, 

SMITH (H. i?!)— The Handbook 
for Midwives. By hbnry Fly 

Smith, With 41 WoodcuU. Cr. 8vo. is. 

SMITH (/. H^—^he White Um- 
brella in Mexico. Bv j. hof- 

' KiNsoN Smith. Fcp. 8vo. 6s. 6</. 
SMITH (R. Bosworfk)-'Cartti^ee 
and the Carthagenians. By r. 

Bosworth Smith, M.A. Maps, Plans, 
&C. Crown 8vo. dr. 

SMITH (S. JK)— Graphics ; or, The 

Art of Calculation by Drawing Lines, 
applied to Mathematics, Theoretical Me* 
chanics, and Engineering, including the 
Kinetics and Dynamics of Machinery, 
&c. By Robert H. Smith, 

Part I. Text, with separate Atlas of 
Plates, 8va 15^ 

SO UTHE K— ThePoeticalWorks 
of Robert Southey. With Por- 

trait. Royal 8vo. 14/. 

STANLEY.—K Familiar History 
of Birds. By £. Stanley, D.D. 
With 160 Woodcuts. Crown 8vo. 3x. td. 

STEEL (/ ^)— Works by. 
A Treatise on the Diseases of 

the Dog I being a Manual of Canine 
Patholc^. Especially adapted for the 
Use of Veterinary Practitioners and 
StndenU. 88 Illustrations. 8vo. los. 6d. 

A Treatise on the Diseases of 

the Ox; beihe a Manual of Bovine 
Pathology specially adapted for the use 
of Veterinary Practitioners and Students. 
2 Plates and 117 Woodcuts. 8vo. 15^. 

STEPHEN.-^RssBYS in Ecclesi- 
astical Biography. By the Right 
Hon. Sir J. Stephen. Cr. 8vo. 7j. 6a. 

STE VENSON{Robt.Louisy-VJoTks 

A Child's Garden of Verses. 

Small fcp. 8vo. 5^. 

The Dynamiter. Fcp. 8va is. swd. 

It. 6d. doth. 

strange Case of Dr. Jekyll and 

Mr. Hyde. Fcp. 8vo. u. swd.; 
\s. 6d. doth. 

The Wrong Box. By Robert 
Louis Stevenson and Lloyd Os- 
BOURNE. Crown 8vo. ^s. 

^r^PCA:— Deductive Logic. By 

St. George Stock. Fcp. 8vo. 3^. 6d. 

'STONEHENGE:—T)it Dog in 
Health and Disease. By 

*Stonehengb.' With 84 Wood En« 
gravings. Square crown 8vo. 7/. 6d. 

STONE Y.^The Theory of the 
Stresses on Girders and 
Similar Structures. By bindo^ 

B. Stonby, LL.D. F.R.S. M.I.C.E. 
With 5 Plates and 143 Illustrations. 
Royal 8vo. 36/. 

Sl/ZZV(/am€sy-VJorks by. 
Outlines of Psychology, with 

Special Reference to the Theory of Edu- 
cation. 8vo. I2s. 6d, 

[Continued on next page. 

CATdvoam or Gkomjx. jam SoMsrmc Booa 

SULLY (/a»Ki)— Works by— 


The Teacher's Handbook of 

Psychology, on the Basb of « Out- 
lines of Psychology.* Cr. 8vo. fir. ttL 

Supernatural Religion ; an In- 
quiry into the Reality of Divine Rere- 
lation. 3 toIs. Sva 36^. 

SWINBURNE.— Victors Logic: 

an Attempt to Popularise the Science ot 
Reasoning. By A. J. SwiMBUEMB, B. A. 
PostSvo. 5a 

THOMPSON (D. Grunieafy- 
Works by. 

The Problem of Evil: an Intro- 
duction to the' Practical Sdenoes. 8yo. 

lOf. ^ 

A System of Psychology. 2 vols. 

8vow 36/. 

The Religious Sentiments of 
the Human Mind. 8vo. 7/. ed. 

Social Progress: an Essay. 8ya 

THORPE.— K Dictionary of Ap- 

?lied Chemistry. By t. e. 
iiORPBf B.Sc (VicL) Ph.D. F.R.S. 
Assisted by Eminent Contributors. 3 
vols. £,z. ax. each. [fW. /. ncm rtady. 

Three in Norway. By Twp of 

Th£1I. With a Map and 59 Illustm- 
tions. Cr. 8vo. ax. bcMurds; ax. &^ doth. 

Times and Days : being Essays in 

Romance and History. Fcp. 8to. 5x. 

TOMSON—Th^ Bird Bride: a 

Volume of Ballads and Sonnets. By 
Graham R. Tomson. Fcp. 8vo. 6x. 

TREVELYAN {Sir G. O. Bart.) 

Works by. 
The Life and Letters of Lord 

Popular Edition, i vol. cr.^. 
Student's Edition, i vol. cr. 8vo. 6x. 
Cabinet Edition, a vols. cr. 8vo. I2x. 
Library Edition, 2 vols. 8vo. 36^. 

The Early History of Charles 

James Fox. library Edition, 8va 
i8x. Cabinet Edition, crown 8vo. 6r. 

TROZLOPE(AntAony)—UoYt]s by. 

The Warden. Crown 8vo.ix.boards; 
IX. 6d. doth. 

Barchester Towers. Crown 8vo. 

IX. boards ; Ix. 6d, doth. 

T WELLS. — Colloquies on 

Preaching. By the Rev. H. 
TWEixs, M.A. Crown 8vo. 5r. 

TYNDALL (/^A«)— Works by. 
Fragments of Sdence. a vols. 

Clown 8vo. lOif. 

Heat a Mode of Motion. Cr. 8vo. 


Sound. With ao4 Woodcuts. 
Ciown 8vo. lox. 6^ 

Researches on Diamagnetism 
and Magne-Crystallic Action. 

With 8 Plates and namerons Ulastntions. 
Crown 8vo. I2r. 

Essays on the Floating-Matter 

of the Air in rektion to PutrefactioD 
and Infection. With 24 Woodcuts. 
Crown 8vo. 7x. ^tU 

Lectures on Light, delivered in 

America in 1872 and 1873. With 57 
Diagrams. Crown 8vo. 5x. 

Lessons in Electricity at the 
Royal Institution, 1875-76. 

With 58 Woodcats. Crown 8vo. ax. 6^. 

Notes of a Course of Seren 
Lectures on Electrical Phe- 
nomena and Theories, down 

8vo. IX. iewed, tx. 6(^ doth. 

Notes of a Course of Nine Lec- 
tures oh Lifl:ht Crown 8vo. ix. 
sewed, ix. 6^ doth. 

Faraday as a Discoverer. Fcp. 

8vo. 3x. 6^ 

CWfWW:— The Testing of Mate- 
rials of Construction: a Text- 

Book for the Engineering Laboratory. 
By W.CawthorneUnwin,F.R.S. With 
5 Plates and 141 Woodcuts. 8fo. atx. 

VIGNOLES—Tht Life of C B. 

VignoleS, F.R.S. Soldier and Civil 
Engineer. By his Son, Olinthus J* 
ViGNOLBS, M.A. 3vo. i6x. 

VILLE.— On Artificial Manures, 

their Chemical Sdection and Scientific 
Application to Agriculture. By Gborgbs 
ViLLB. Translated and edited by W. 
Ckookes. VTith 31 Plates. 8vo. atx. 

VIRGIL— FMhW VergiU Maronis 
Bucolica, Georgica, ^Sneis; 

the Works of Viroil, Latin Text, with 
English Commentary and Index. By 
B. H. KSNNEDY, D.D. Cr. 8vo. tox.M 




The /Eneid of Virgil. Translated 
into English Verse. By John Coning- 
TON| M.A. Crown 8to. 6r« 

The Poems of VIrgiL Translated 
into English Proie. By John Coning- 
TON, M.A. Crown 8vo. 6x. 

WALKER,— T\i!t Correct Card; 

or» How to Play at Whist; a Whist 
Catechism. By Major A. Campbell- 
Walur, F.R.G.S. Fq). 8vo. 2s. 6^ 

»^^Z/>(9Zi?.— HistoryofEnrfand 
from the Conclusion of the 
Great War in z8i$ to 1858. 

BySPENCBaWALFOLB. Library Edition. 

5 vols. 8va £\, lOf. Cabinet Edition, 

6 vols, crown 8vo. 6f. each. 

WATERS.— -Pzx^^ Registers in 

Enfi^and : their History and Contents. 
By K. £• Chester Waters, B.A. 
8vo. 5J. 

WATTS' Dictionaryof Chemistry 

Revised and entirely Re-written by H. 
FOR5TER MoRLEY, M.A. D.Sc. ; and 
M. M. Pattison Muir, M.A. F.R.S.E. 
Assisted by Eminent Contributors. To 
be published in 4 vols. 8vo. Vols. I. & II. 
A— Indigo. 42J. each. 

fTE^^.— Celestial Objects for 
Common Telescopes. By the 

Rev. T. W. Webb. Crown 8vo, qs. 
WELLINGTON. — Ui^ of the 
Duke of Wellington. By the 

Rev, G. R. Gleio, M.A. Cr. 8vo. y. 6d. 

WELLS. — Recent Economic 

Changes and their Efiect on the Pro- 
dnction and Distribution of Wealth and 
the Well-being of Society. By David A. 
Wells, LL.D. D.C.L. late United States 
Special Commissioner of Revenue, &c. 
Crown 8vo. lOf. ^ 

WENDT.—PsipeTS on Maritime 

Legislation, with a Translation of 
the German Mercantile Laws relating to 
Maritime Commerce. By Ebnest Emil 
Wbndt, D.CL. Royal 8vo. £1. lis. 6d. 

WEST {Charles)— \rj0T\LS by. 

Lectures on the Diseases of In- 
fancy and Childhood 8vo. iSx. 

The Mother's Manual of Chil- 
dren's Diseases. Cr. 8vo. 2/. 6d. 
WEYMAN—Th^ House of the 

Wolf: a Romance. By Stanley J 
Weyman. Crown 8vo. 6/. 

WHATEL Y(E.Jane.)—^ox^ by. 

English Sjrnonyms. Edited by R. 
Whately, D.D. Fcp. 8vo. y. 

Life and Correspondence of 
Richard Whately, D.D. ute 

Archbishop of Dublin. With PortraiU 
Crown 8vo. lOf. 6^ 

WHATELY {ArMisAo/)— Works 

Elements of Logic. Cr. 8vo. 4s. 6d. 
Elements of Rhetoric. Crown 

8vo. 4r. 6J. 

Lessons on Reasoning. Fcp.8vo. 

IS. td. 

Bacon's Essays, with Annotations. 

8vo. lOf. 6^. 

WILC0CKS.—Tht Sea Fisher- 
man. Comprising the Chief Methods of 
Hook and Line Fishing in the British and 
other Seas, and Remarks on Nets, Boats, 
and Boating. By J. C. WiLCOCKS. 
Profusely Illustrated. Crown 8vo. 6x. 

»7Z^^.— Lectures on Patholog- 
ical Anatomy. Bv Samuel WiLKs, 
M.D. F.R.S. and the late Walter 
Moxon, M.D. F.R.C.P. 8vo. i8j. 

WILLIAMS.— VvAmonssj Con- 
sumption; its Etiology, Pathology, 
and Treatment. By C. J. B. Williams, 
M.D. and Chas. Theodorb Williams, 
M.A. M.D. Oxon. With 4. Coloured 
Plates and 10 Woodcuts. 8vo. i6f. 

WILLIAMS.— llL^xiM^X of Tele- 
graphy. By W. WiLLLOCS, Superin- 
tendent of Indian Government Telegraphs. 
Illustrated by 93 Engravings. 8vo. iQr . 6^. 

WILLIAMS— Th^ Land of My 

Fathers : a story of Life in Wales. 
By T. Marchant Williams. Cr. 8vo. 
2s. 6d. 

WILLICB:— Popular Tables for 

giving Information for ascertaining the 
value of lifehold. Leasehold, and Church 
Property, the Public Funds, &c. By 
Charles M. Willich. Edited by 
H. Bbnce Jones. Crown 8vo. lOi. 6d. 

WILLOUGHB Y. — East Africa 

and its Big Game. The Narra- 
tive of a Sporting Trip from Zanzit)ar to 
the Borders of the Masai By Capt. Sir 
John C. Willoughby, Bart. Illus- 
trated by G. D. Giles and Mrs. Gordon 
Hake. Royal 8vo. 2IJ. 

Catalogue of GBimRAL ajjd SciENTtFtc Books, 

WILSON.— h Manual of Health- 
Science. Adapted for Use in Schools. 
By Andrew Wilson, F.R.S.E. F.L.S. 
&C. Witb74llIustnUions. Cf.Sto. 2/.6^ 

»Yrr(/Vi?/)— Works by. Trans- 
lated by Francis Younghusbamd. 

The Trojan War. Crown 8vo. 2 j. 

Myths of Hellas; or, Greek Tales. 

Crown 8vo. 3/. 6^. 

The Wanderings of Ulysses. 

Crown 8va 51. 

^(^Z/^^.-'Rambles in tht Black 

Forest By Henry W. Wolff. 
Crown 8vo. 7j. dd, 

WOOD {Rev. J. a)— Works by. 

Homes Without Hands ; a De- 
scription of the Habitations of Animals, 
classed according to the Principle of Con- 
struction. With 140 Illustrations. 8vo. 
lor. 6<r. 

Insects at Home ; a Popular 

Account of British Insects, their Struc- 
ture, Habits, and Transformations. With 
700 Illustrations. 8vo. lOf. fxi. 

Insects Abroad ; a Popular Account 

of Foreign Insects, their Structure, 
Habits, and TransTormatioDS. With 
600 Illustrations. 8vo. lor. 6J. 

Bible Animals; a Description of 

every Living Creature mentioned in the 
Scriptures. With lis Illustrations. 8vo. 
lOf. 6^. 

Strange Dwellings ; a Description 

of the Habitations of Animals, abridged 
from 'Homes without Hands.* With 
60 Illustrations. Crown 8yo. 3x. 6^. 

Out of Doors; a Selection of 

Original Articles on Practical Natural 
History. With ii Illustrations. Crown 
8vo. 3J. 6^. 

Petland Revisited. With 33 

Illustrations. Crown 8vo. y, 6d. 

The following books are extracted from the 
foregoing works by the Rev. J. G. Wood : 

Social Habitations and Parasitic 

Nests. With 18 Illustrations. Crown 
8to. 2X. 

The Branch Builders. With 28 

Illustrations. Crown 8vo. ar. 6d. 

WOOD {Rev. J. C.)— Works by- 


Wfld Animals of the Bible. 

With 29 lUustntions. Crown 8to. 3r. 6i/. 

Domestic Animals of the Bible. 

With33Hlasti»tions. Crown 8to. 3/. 6^. 

Bird-Life of the Bible. With 32 

lUustntions. Crown 8Ta y. 6^. 

Wonderful Nests. With 30 Illus- 
trations. Crown 8yo. y. 6d. 

Homes under the Ground. With 

28 IllustnUions. Crown 8vo. 3r. &f. 

YOUATT {Wi/iiam)^'Works by. 

The Horse. Revised and enlarged 
8vo. Woodcuts, 7J. 6rf. 

The Dog. Revised and enlarged. 

8vo. Woodcuts, dr. 

Works l^. 

The Story of our Lord, told 
in Simple Language for 

Children. With 25 illustrations oa 
Wood from Pictures by the Old Masters. 
Crown 8vo. ar . 6<^ 

The Story of Genesis. Crown 8vo. 

ZELLER {Dr. -£.)— Works by. 
History of Eclecticism in Greek 

Philosophy. Translated by Sarah 
F. Allbynb. Crown 8vo. lor. ^, 

The Stoics, Epicureans, and 

Sceptics. Translated by the Rev. 
O. J. RaiCHEL, M.A. Crown 8vo. 15J. 

Socrates and the Socratic 

Schools. Translated by the Rev. O. 
J. Rrichkl, M. a. Crown 8vo. lOr. 6d. 

Plato and the Older Academy. 

Translated by Sarah F. Alletnk and 
ALf RBD Goodwin, B.Al Crown 8vo. 

The Pre-Socratic Schools: a His- 

tory of Greek Philosophy from the Earliest 
Period to the time of Socrates. Trans- 
Uted bv Sarah F. Allbynb. a vols, 
crown ovo. 301. 

Outlines of the History of 
Greek Philosophy. Trans ^ 

by Sarah F. Allbynb and £vi /H 
Abbott. Crown 8vo. lor. 6^, 

Spottasyroodc & Co. Printers, New-street Square, London. 

14 DAY USE 'J7^ 


Ttu. . 

on UM diBt!0 to whidi fco6wc!<L 

E«oew«d twdci ftc« i«bjccr to uotnediMe tvcmIL 

W^V.jU I3G2 


DEC 2 2 19G9 

uc J^»»*^ 





*NOSd3id "H n ONv Nosaeid saw ^o idio 

;isiosH3id w wvmiM 

dO JlHVtfflU 3HJ, ViOHd 


viNHO^nvo AO AiisaaAiNfi 

. \ ^ - ^ ^ ^ "" ^ ■ ■